A research group from the National Cancer Institute is trying to outlicense a SNP-based diagnostic that it claims can help predict side effect response to paclitaxel, a widely prescribed chemotherapeutic.
The William Figg lab at NCI’s Center for Cancer Research is looking to license the technology, which is based on three SNPs in the ABCB1 gene that are linked to a range of drug-related side effects.
In a pilot study now in press in the European Journal of Cancer, the researchers write that ABCB1 polymorphisms are associated with neutropenia, neurotoxicity, and myelotoxicity in cancer patients being treated with paclitaxel.
The study was supported by Bristol Myers-Squibb, which originally launched the drug under the brand name Taxol.
The lab has filed a patent application for this finding and is now looking for a partner to develop the technology into a marketable SNP-chip assay.
“We are just now determining who [potential] partners might be in the pharmacogenetics world,” said Figg, a senior scientist and section head of Molecular Pharmacology at NCI.
According to the NCI, the test would be given before chemotherapy with paclitaxel so high-risk patients could begiven fewer treatment cycles, monitored more closely for adverse effects, or potentially switched to other chemotherapies with different side-effect profiles.
While there aren’t yet many such products on the market for identifying patients at risk for developing adverse events while undergoing chemotherapy, “this is the crack in the dam,” said Sam Tetlow, a principal with Clearview Limited, a consulting service in Raleigh, NC.
Eventually, he said, “this [market] can get pretty big. Today, the molecular diagnostics market is hugely dominated by infectious disease — HIV and AIDS — but cancer is probably a big one, perhaps behind neuropsychiatric disorders.”
ABCs of AEs
ABCB1, also known as the multidrug-resistance protein (MDR-1) and P-glycoprotein, helps distribute and eliminate paclitaxel from the body. Cells in the blood-nerve barrier express ABCB1, and the protein may be involved in protecting the peripheral nervous system by transporting potential toxins away from nervous system cells and into the bloodstream.
Three well-characterized and common SNPs in the ABCB1 gene, 1236C>T, 2677G>T/A, and 3435C>T, have been shown to alter expression of this protein. In short, the polymorphisms may cause ABCB1 to be differentially expressed in the blood-nerve barrier and in precursor hematopoietic cells.
This is noteworthy because, for example, the frequency of the C3435 allele ranges from about 34 percent to as much as 90 percent in some populations.
“We identified that Taxol was transported by the ABCB gene, identified the fact that it’s expressed on nerve cells, and thus we could predict who would developed side effects, based on the presence or absence of a functional transporter,” said Figg.
The success of paclitaxel therapy is often limited by neuropathy. In vitro work shows the drug inhibits neurite growth, damages neurons and glia, and leads to demyelination.
Clinically, it often causes mild to moderate peripheral numbness, but sometimes these symptoms are severe: Between 9 and 11 percent of women with metastatic breast cancer treated with paclitaxel developed grade III neuropathy, meaning functionally disabling numbness, complete loss of reflexes, and generalized weakness. Neutropenia and susceptibility to infection are also common problems.
A diagnostic that could help identify patients at risk for such adverse events might find a large market. Paclitaxel, frequently prescribed in combination with cisplatin, is approved for advanced ovarian cancer and non-small-cell lung cancer, and as an adjuvant therapy for node-positive breast cancer, metastatic breast cancer, and Kaposi’s sarcoma.
It is being investigated for use in many other cancers, including advanced head and neck cancer, early-stage ovarian epithelial cancer, endometrial cancer, and metastatic prostate cancer.
Bristol Myers Squibb said global Taxol sales declined 25 percent in 2005 to $747 million. The company lost Taxol patent exclusivity in 2000 in the US and in 2003 in the EU.
Asked if BMS might be a potential partner, Figg said that the drug company had not yet been approached, but thought that it was not a likely candidate. A BMS spokesperson said that he could not comment.
One similar test already on the market is Third Wave’s Invader UGT1A1 molecular assay, which detects a common polymorphism associated with severe side effects caused by Pfizer’s colorectal cancer drug Camptosar, known generically as irinotecan.
The US Food and Drug Administration approved Third Wave’s test last year, and, importantly, mandated that the drug’s label be changed to recommend treatment alterations for those with the relevant allele. A Third Wave spokesperson said that the product has not yet contributed in a meaningful way to the company’s revenue.
Clearview Limited’s Tetlow also mentioned Genomic Health’s Oncotype Dx, which predicts response to tamoxifen and suggests the likelihood that early-stage breast cancer will recur.
While there aren’t yet many such products on the market for identifying patients at risk for developing adverse events while undergoing chemotherapy, “this is the crack in the dam.”
Not only can this sort of diagnostic save patients from unnecessary treatment, said Tetlow, but it may save money. “Insurance companies see huge benefits. Any woman diagnosed with breast cancer five years ago got $20,000 worth of tamoxifen, and some don’t need that much.”
Despite these bright spots, challenges remain. For example, clinical standards for a toxicity diagnostic are very high in the oncology sphere, cautioned Keith Batchelder, CEO of consultancy Genomic Healthcare Strategies. In cancer treatment, “you have to have excruciatingly high sensitivity and specificity [of a test] before you alter the course of therapy you give to the patient,” he said. “I hate to put it this way, but if the alternative is potential death, you’ll probably put up with some neurotoxicity.”
Batchelder said that chronic conditions such as asthma or arthritis may be more suitable for molecular diagnostics that are focused on side effects, since the clinical utility of such tests can be more obvious to patients, physicians, regulators, and insurers.
Identifying the role of these SNPs in patient response is important in understanding the underlying biology, but may not be as directly useful in developing a marketable diagnostic, said Batchelder. “Until they’re validated with a lot of patient data and can be tied to how they change the course of therapy, the quality of care, and the cost of care, it’s going to be a hard thing to develop.”
Figg said that in seeking a partner, his lab hoped to further validate these SNPs with more patient data. “We need DNA and outcome information, and we’ve exhausted what we have here,” he said.
Another challenge is that the regulatory process for these diagnostics has not yet been sorted out. The FDA recently released new draft guidance for multiplexed assays in an attempt to clarify its expectations, but “the regulatory requirements are very much in a state of flux,” said Tetlow [see 9/13/2006 PGx Reporter]. “Companies are by and large working with FDA to follow either a homebrew or a 510(k) strategy — the FDA is basically playing catch-up, as they often have to.”