Skip to main content
Premium Trial:

Request an Annual Quote

Roche Looks to Take Lead in Emerging Theranostics Space With p53 AmpliChip


SAN FRANCISCO — Roche Diagnostics is taking advantage of the US Food and Drug Administration's nascent drug-diagnostic co-development "concept paper" by moving a companion diagnostic-therapeutic cancer product through its pipeline, the Swiss company said last week.

Using the existing co-development framework released by the FDA last month [see Pharmacogenomics Reporter 4/14/2005], the diagnostics giant will "absolutely" develop its P53 AmpliChip resequencing assay, which runs on Affymetrix's GeneChip reader, together with a cancer drug known tentatively as Nutlin, Jim Gallarda, director of marketing development of clinical genomics at Roche, told Pharmacogenomics Reporter between sessions at the Cambridge Healthtech Institute's Molecular Medicine Tri-Conference, held here last week. Gallarda's group is largely responsible for AmpliChip's entry into the market.

The co-development, if successful, will result in a cancer drug accompanied by a test capable of identifying patients with solid tumors for which the drug likely will not work. Roche would not disclose the specific indication for which it would apply the chip or its therapeutic mate. With Nutlin and P53, Roche is taking an early lead in the co-development process, following up on its success with AmpliChip P450, which became the first microarray product to achieve FDA clearance for use as an in vitro device [see PGx Reporter 3/24/2005].

Roche is already using the test to stratify patients during clinical trials of Nutlin, and it is planning on partnerships with others in pharma. However, Roche plans to investigate whether the chip can be useful for cancer patient prognosis or in identifying patients who may respond to other particular classes of drugs, Gallarda said.

Roche has not yet filed the AmpliChip P53 for clearance with the FDA, and the company has not decided whether it would take the 510(k) or premarket clearance route, said Gallarda. It is faster to achieve 510(k) clearance, while a PMA "allows you to make more specific claims about the intended use of the test," he said. The AmpliChip P53 will not be released prior to FDA clearance, he added.

He said Roche will move the chip into trials in 2006, with commercialization dependent on the outcome. "We're exploring several different areas of cancer for the potential value of a P53 diagnostic," although bladder and lung cancer "look pretty promising" for stand-alone IVD chip indications, he said. "I predict the P53 product's availability for [research] diagnostic use before the drug is approved by the FDA," Gallarda added.

The company has also not yet decided how to label the co-developed products in case they are approved. "How do [you] label the device when a drug is affected, and how do you label the drug when [a] device is required?" asked Gallarda rhetorically.

Assessing the P53 Marker

Until Roche decides on an indication, determining the potential markets for the drug and the diagnostic will be difficult. About half of all human cancers are associated with mutated P53. Nutlin is an imidizoline-class compound that inhibits the binding of the MDM2 protein, which is an inhibitor of the P53 protein.

The P53 protein acts through two mechanisms to fight cancer — which of the two predominates depends on the tissue type, according to Andrei Gudkov, chairman of the Department of Molecular Genetics at the Lerner Research Institute of the Cleveland Clinic Foundation. In lymphatic or hematopoietic tissues, for example, the protein tends to produce apoptosis in pre-cancerous cells while it promotes growth arrest in response to stress (such as chemo- or radiation therapy) in tissues of epithelial origin. Bladder and lung are epithelial tissues.

The next AmpliChip in Roche's assembly line after p53 is a leukemia-typing expression array, Gallarda said.

Investment by pharma into pharmacogenomics is about twice as high in the field of oncology as in the next field in line, neurology, according to a Roche analysis cited by Angus Hastie, director of global marketing and business development in clinical genomics at Roche.

In a presentation at the Tri-Conference last week featuring the P53 chip and Nutlin, Ian Massey, Roche's head of research and preclinical development, focused on P53's apoptotic pathway. In this mechanism, cells with wild-type P53 experience a build-up of the protein that results in apoptosis. In many tumors, duplication or up-regulation of MDM2 suppresses P53 and allows continued growth. Mutated, malfunctioning P53 and other pathway defects can also produce the same effect.

Using the P53 AmpliChip to support an MDM2 binding inhibitor "makes complete sense because [when] we have the therapy … we [will] really need to know P53 status" for patient stratification, said Gudkov. "This diagnostic assay is essential for this type of focused, targeted treatment."

Howard McLeod, a professor of medicine at Washington University, describes himself as a P53 cynic, though he explicitly praised Roche's microarray technology. The MDM2 inhibitors may "end up being great," but the drugs are still too early in clinical trials to assess whether they make good drugs, he told PGx Reporter. "Almost every major pharma has had a program in that area, so it could be that there are a ton of them," he said. "If there are a few out there in the public domain, there are probably a hundred that are still being evaluated."

There are at least three MDM2 inhibitors in the Nutlin family, and at least one more MDM2 inhibitor goes by the name of RITA.

Because P53 is such an important player in cancer, Roche may hope its new chip will find broad use in clinical trials. "We're in discussions with other pharma companies on the potential of the use of [the] P53 sequencing [chip] in their drug development programs," Gallarda said. Both large and small drug makers have shown interest in co-developing with the chip, he said, though he declined to elaborate. "We're establishing several drug exploratory collaborations that will be of real benefit to looking at the utility of the test," he added.

As a prognostic tool, on the other hand, immediate uses for the P53 chip are less obvious. "With new drugs that are specific for these pathways, it may have dramatic impact," said McLeod. "'Show me the data,' basically. If there is some data saying that there is a relationship between P53 response and any therapy, great. But at the moment, the data that is in the public domain, anyway, it's not really very relevant," he said.

No existing chemotherapy has proved to be linked to P53 status, he said. Indeed, the chip will only be useful as a tool to stratify patients in clinical trials involving drugs directed at the action of P53, such as MDM2 inhibitors, said Jorge Leon, head of Leomics, a Princeton, NJ, molecular-diagnostics consulting firm. "There will be companies developing drugs where it is important to know P53 status, but not a lot," he said. But the co-development takes good advantage of the company's capabilities, and will serve as a model for followers, he added.


Although P53 has long been the subject of prognosis studies, it hasn't yet entered clinical practice. "The reason that [P53 genotyping] hasn't taken off is that it's still sort of unclear whether [the] P53 sequence means anything" outside of cell lines and mice, Washington University's McLeod said. He cited a literature review conducted with colleagues containing "about 3,500" patients' worth of data. Ultimately, no prognosis could be reliably gleaned from P53 sequence — the gene's mutational status had an effect on outcome "about half the time," with the remainder of studies reporting no effect, said McLeod.

"I'm not aware of a single critical procedure which would depend on the knowledge of whether p53 is wild-type or mutant," although the chip may conceivably lead to that situation through its use by academic researchers and clinical trials at pharmaceutical companies, said Gudkov. "There are plenty of things cooking, though," he added.

— CW

Filed under

The Scan

Study Links Genetic Risk for ADHD With Alzheimer's Disease

A higher polygenic risk score for attention-deficit/hyperactivity disorder is also linked to cognitive decline and Alzheimer's disease, a new study in Molecular Psychiatry finds.

Study Offers Insights Into Role of Structural Variants in Cancer

A new study in Nature using cell lines shows that structural variants can enable oncogene activation.

Computer Model Uses Genetics, Health Data to Predict Mental Disorders

A new model in JAMA Psychiatry finds combining genetic and health record data can predict a mental disorder diagnosis before one is made clinically.

Study Tracks Off-Target Gene Edits Linked to Epigenetic Features

Using machine learning, researchers characterize in BMC Genomics the potential off-target effects of 19 computed or experimentally determined epigenetic features during CRISPR-Cas9 editing.