The Test and the Target: Diagnostics’ Critical Role in Individualized Cancer Care
In what are still the early days of personalized cancer medicine, much of the focus has been on the use of molecularly targeted therapies, such as imatinib (Gleevec) or trastuzumab (Herceptin). But this is beginning to change as cancer researchers and pharmaceutical and biotechnology companies devote greater attention to an integral part of the therapeutic equation: the tests—or “companion diagnostics,” as they are often called—needed to determine not only whether a patient’s tumor expresses the molecule being targeted, but also whether that expression correlates with the patient’s response to the treatment.
Assistance for Diagnostics Developers
Based on a 2007 workshop sponsored by NCI and the FDA, a recent commentary published in the Journal of the National Cancer Institute, for which Dr. Taube was the lead author, laid out the challenges for biomarker development, including developing companion diagnostics in concert with a molecularly targeted agent.
Building on that earlier conference, as part of the PACCT initiative, NCI is launching a clinical assay development program that will establish a network of laboratories, including an NCI laboratory, to help both industry and academic researchers develop validated, clinical-grade companion diagnostics, Dr. Taube explained. This will complement and build upon the efforts that NCI initiated a number of years ago to develop standard operating practice (SOP)-driven assays to measure pharmacodynamic biomarkers in Phase 0 and I clinical trials.
“Our initial focus will be on tests that are proposed as integral assays that must be done in order for a trial to proceed,” she said. That would include, for example, an assay to determine patient eligibility for the trial or to assign patients to a specific treatment arm.
The effort, which is just beginning and being supported with funds from the American Recovery and Reinvestment Act, will initially focus on assays for phase III trials and large phase II trials. More information on the program will be available in the coming months.
It’s all part of the move away from “nonselective therapeutics,” explained Dr. Paul Mischel, a researcher at UCLA’s Jonsson Comprehensive Cancer Center. “If we’re dealing with therapies that target specific enzymes, the alterations in those enzymes and the pathways that they regulate are often different in patients with the same types of cancers, so having companion diagnostics is essential” to individualizing patient care, he said.
Companion diagnostics are intimately tied to the development of diagnostic and prognostic biomarkers; novel clinical trial designs, such as so-called adaptive design trials; and initiatives like NCI’s Cancer Human Biobank to standardize and centralize tissue sample procurement and storage—all part of a collective effort to spur more rapid advances in personalized cancer treatment.
All of these approaches, argued Drs. Richard Schilsky, Anil Potti, and Joseph Nevins in a recent commentary in Science Translational Medicine, are “critical for a renewed war on cancer: Without a robust mechanism for selecting the right treatment for the right patient at the right time, we will continue to see only incremental improvements and have little hope for substantial survival gains.”
Complexity Fueling Collaboration
The difficulty and expense of developing a successful cancer treatment have been well documented. But the challenge of developing diagnostics that are, in effect, therapeutic partners for a given treatment, “can’t be understated,” stressed Dr. Mischel, whose laboratory, in collaboration with Dr. James Heath from the California Institute of Technology, is focused on using nanotechnology and other platforms to develop diagnostics that can predict response to therapy and be used to noninvasively monitor response after treatment has begun. “Creating the tests, getting them into [Clinical Laboratory Improvement Amendment]-certified labs, creating testing kits, and being able to do these things in a way that is highly standardized and highly reproducible is just a very difficult thing to do,” he said.
The technological and logistical aspects are not the only difficulty. An additional challenge comes in trying to measure the right thing, said Dr. Sheila Taube from NCI’s Program for the Assessment of Clinical Cancer Tests (PACCT) initiative. (See sidebar.)
“Cancer biology is very complex, and there are numerous interconnected pathways in a cancer cell,” she said. “If you cut off one pathway, there may be another pathway the cell uses to accomplish the same end. So if you measure only one piece of a pathway that may be involved, that may not be sufficient to know whether the patient will respond to the drug.”
Both the increased understanding of cancer’s complexity and the complexity of developing companion diagnostics are driving many pharmaceutical companies’ activities in the area of personalized medicine, said Dr. Cecilia Schott, AstraZeneca’s business development director for personalized healthcare. Part of that change is the growing number of companies that are establishing partnerships with diagnostics firms, as AstraZeneca did earlier this year in inking a deal with Dako Denmark.
“It’s challenging to find the biomarker, determine whether it is meaningful, and then to identify the right technology to measure it,” Dr. Schott said. So, at earlier stages in the drug development process, there is a greater emphasis on understanding the biological importance of a biomarker in certain tumor types, as well as identifying the best way to measure that marker. As for the testing platforms, she continued, “You may have to try several in parallel until you find the one that is best to use.”
Although all of these efforts may make the drug development process more complicated early on, the long-term pay-off, in terms of success of a drug and impact on patients, will likely be greater, Dr. Schott said.
Developing a Drug and a Test in Unison
“We were thinking about it from the beginning,” explained Dr. Peter Hirth, CEO of Plexxikon—the “it” referring to a companion test to identify the best candidates for the company’s investigational drug PLX4032. The agent targets a specific mutation in the BRAF gene, which is present in approximately 70 percent of patients with melanoma. In a phase I clinical trial, the response rate to treatment with PLX4032 (at least a 30 percent reduction in tumor size) was 70 percent in patients with the mutation, including the complete eradication of tumors in some patients.
Although the phase I dose-escalation trial was open to any patient with metastatic melanoma, an “extension cohort” was later opened, enrolling only patients whose tumors had the BRAF mutation, assessed using a PCR-based test developed by Roche, which is an official collaborator with Plexxikon in the development of PLX4032. This test is also being used to enroll patients in a phase II trial, as well as a recently launched phase III trial meant to establish PLX4032 as a first-line treatment for patients with advanced melanoma who have the BRAF mutation.
Should the phase III trial yield positive results, the specifics on how to perform the test and analyze the results, based on work done as part of the clinical trials, will be part of the overall submission package to the FDA for approval to market the drug, Dr. Hirth noted.
To the degree that your drug program allows it, it’s never too early to start thinking about developing a companion diagnostic.
—Dr. Elizabeth Mansfield,
Food and Drug Administration
Pfizer has taken a similar route with several of the cancer drugs in its development pipeline, including crizotinib (formerly known as PF-02341066), which targets the ALK gene and has shown promising results in a phase I trial in patients with advanced lung cancer. The phase I trial was enriched with individuals whose tumor cells had a fusion of two genes, including ALK. The patients were identified using a FISH-based assay developed by Abbott, under an agreement between the two companies. According to a Pfizer spokesperson, the assay will be used to enroll patients in a recently launched phase III trial of the drug.
This start-early approach is highly advisable, said Dr. Elizabeth Mansfield from the Office of In Vitro Diagnostic Device Evaluation and Safety in the FDA’s Center for Devices and Radiological Health (CDRH). “To the degree that your drug program allows it, it’s never too early to start thinking about developing a companion diagnostic,” she said.
And the FDA is doing what it can to help researchers and companies move things in that direction. “If there is even an inkling that there is a potential for targeting a specific population [with an investigational drug], even if hasn’t been tested in humans yet…we encourage them to get in contact with CDRH,” said Dr. Issam Zineh, from the Office of Clinical Pharmacology in the agency’s Center for Drug Evaluation and Research.
There’s been little doubt or confusion about the need for companion diagnostics to individualize cancer therapy, said Dr. Mischel. “But what’s different now,” he continued, “is that the business models are catching up” with that understanding.