By Turna Ray
NanoString is developing a gene expression test to help oncologists and pathologists determine the risk of cancer relapse and guide treatment based on intrinsic subtypes — luminal A, luminal B, HER2-enriched, basal-like, and normal-like.
NanoString CEO Brad Gray told PGx Reporter this week that unlike other genomic breast cancer tests on the market, the company is betting that its test will be able to guide treatment with specific classes of chemotherapies, such as taxanes and anthracyclines.
Under an agreement announced last week, NanoString gained the rights to a 50-gene signature, dubbed PAM50, from Bioclassifier, a company founded by four breast cancer experts from four research institutions — University of North Carolina at Chapel Hill, Washington University, University of Utah, and the BC Cancer Agency.
NanoString is developing a test based on the panel that will run on its nCounter multiplex gene expression platform. The firm is planning to gain regulatory approval from the US Food and Drug Administration for the platform and the PAM50 panel and hopes to have the test on the market by 2012.
Currently, the molecular testing market for breast cancer recurrence is led by Genomic Health's Oncotype DX and Agendia's MammaPrint. These tests gauge whether women are at high risk of recurrence and therefore need to be treated with chemotherapy plus hormonal therapy, or are at low risk of recurrence and can be treated with just hormonal therapy (tamoxifen).
Genomic Health's Oncotype DX has been on the market since 2004. Operating on an RT-PCR platform, the test assesses the expression of 21 genes in a patient's tumor sample and yields a recurrence score that translates into high risk, low risk, or intermediate risk for patients. The 21-gene recurrence score also corresponds to the likelihood that a woman will benefit from the addition of anthracycline-based chemotherapy — cyclophosphamide, doxorubicin, fluorouracil — plus tamoxifen. While the test is most readily used for disease prognosis and treatment benefit prediction in newly diagnosed, stage I or II, node-negative, estrogen receptor-positive breast cancer patients, Oncotype DX is also being marketed now in node-positive patients.
Meanwhile, Agendia's MammaPrint — a breast cancer recurrence test cleared by the US Food and Drug Administration as an in vitro diagnostic multivariate index assay in 2007 — uses a microarray platform to gauge the expression of 70 genes. The test has been validated in early-stage breast cancer patients at risk of recurrence following surgery, despite ER status and any prior treatment. Depending on whether the test yields a high or low risk of distant recurrence, a doctor can decide whether the patient should be treated with just tamoxifen, or needs chemotherapy in addition to hormonal therapy.
Also, last week Clarient launched Mammostrat, an immunohistochemistry-based test that gauges five biomarkers to assess a woman's risk of breast cancer recurrence following surgery and initial treatment. Mammostrat is intended for use in postmenopausal, node-negative, estrogen receptor-positive breast cancer patients who will receive hormonal therapy and are considering adjuvant chemotherapy (PGx Reporter 07/14/10).
The intrinsic subtyping assay that NanoString is developing based on the PAM50 signature will improve upon existing tests because it will gauge the risk of breast cancer recurrence "across all stages of breast cancer and across all women who present with ER-positive and -negative [disease]," Gray told PGx Reporter. "So, there are women in the market who we believe we can help by intrinsic subtyping, who have not been helped by some of the other tests on the market."
Gray added that the assay will not only be able to determine whether women have a low risk of recurrence and should therefore be treated only with tamoxifen, but will ideally also be able to predict which subclasses of women will or won't respond to anthrycyclines and taxanes.
Although NanoString is entering a somewhat crowded diagnostic testing market, there is a substantial body of evidence on the ability of the PAM50 gene signature to predict disease prognosis and treatment benefit.
For example, a 2009 study published in the Journal of Clinical Oncology reported that the 50-gene subtype predictor would be "of value for the management of node-negative breast cancers." Among the authors in this study were the four cofounders of Bioclassifier and developers of the PAM50 signature: Charles Perou of the University of North Carolina at Chapel Hill; Matthew Ellis of the Washington University School of Medicine; Philip Bernard of the University of Utah and Huntsman Cancer Institute; and Torsten Nielsen of the BC Cancer Agency.
[ pagebreak ]
In the study, the researchers analyzed archival formalin-fixed, paraffin-embedded tissue samples with the 50-gene subtype predictor using Agilent's microarrays, as well as with qRT-PCR. The aim of the study was to look at the utility of intrinsic subtypes in patients receiving no adjuvant systemic therapy, as well as in those being treated with paclitaxel/CAF neoadjuvant chemotherapy. The authors ultimately concluded that "the subtypes and risk score can … be used to assess the likelihood of efficacy from neoadjuvant chemotherapy" with a taxane and anthracycline regimen.
In a follow-up study published later that year in Cancer Research, several of the same authors evaluated the ability of the PAM50 subtypes to predict pathological complete response to anthracycline/taxane-based chemotherapy in the neoadjuvant setting. In a comparison against ER status, PR status, HER2 status, node status, and tumor size, the intrinsic molecular subtype classification was found to be "the most significant predictor of [pathological complete response] across three anthracycline/taxane-based neoadjuvant cohorts," the authors wrote.
Eyeing the POC Market
NanoString plans to run the PAM50 test on its nCounter platform, which it believes offers advantages over PCR and microarrays for use in the point-of-care market.
While firms like Genomic Health and Agendia analyze patient samples at a centralized lab, NanoString's vision is to market its breast cancer test in a decentralized manner, bringing it closer to the patients being treated, at hospitals and in pathologists' labs.
Gray estimated that the PAM50 test run on the nCounter platform would yield test results in 48 hours and keep the patient sample within the treatment facility. By using a point-of-care test, "the relationship between the local pathologist and oncologist is preserved," Gray said. "The pathologist can play the role that they have traditionally played, providing counsel to the oncologists … about the characterization of the tumor types and the information about the right treatment for the patient."
Describing the test as "incredibly simple," Gray noted that the 50 genes can be measured in a single tube, and the test requires 15 minutes of hands-on time. "The rest of the process is automated on the liquid handling robot and a scanner," he said. "The technology … is simple enough that we feel the medical technologist can perform it in a clinical laboratory."
Furthermore, the test does not require an amplification step, and relies on digital counting of individual molecules to measure expression. "So, even in a distributed model, we can have a very simple and reproducible assay, ensuring that women will get the same answer about their intrinsic subtype … regardless of where that test is performed," Gray noted.
A test that can be performed at multiple laboratories with high reproducibility would also "ensure that people not just in the US but all over the globe have access to the latest innovations in molecular diagnostics," Gray said. "That's a challenge that centralized labs are trying to meet by shipping samples from all over the world into the US. But it's intuitive that global access is much easier to provision in a distributed way than in a centralized way."
Gray acknowledged that NanoString will face challenges when it comes time to seek FDA approval for the test, but stressed that the agency's approval will be necessary in order to pursue the POC model.
For example, the amount of data that NanoString will have to provide the agency regarding the analytical and clinical validity of the test will be much greater that what a CLIA laboratory typically has to prepare for a laboratory-developed test, "but I think we're lucky that we have a platform with attributes that are naturally suited to a distributed diagnostics model," he said.
"This is a slower and far more expensive [path], than setting up a service laboratory," Gray explained. "There is a tremendous level of analytical validity studies and clinical validity studies that need to be performed in advance of seeing a dollar of diagnostic revenue. That presents, for all companies, a challenge from a business perspective."
Gray said it is still too early to say whether NanoString will file for 510(k) or premarket approval. Neither the PAM50 panel nor the nCounter platform has undergone FDA review.
Given the regulatory hurdles, NanoString is targeting a 2012 launch date for the nCounter-based PAM50 test. In the meantime, ARUP Laboratories, a wholly owned subsidiary of the University of Utah, will launch a PCR-based LDT based on the same gene signature in about a month, according to PGx Reporter sister publication PCR Insider.
Since the commercial rights to the PAM50 signature are owned by Bioclassifier, each of the founding researchers' academic centers have rights to use the signature for research or commercial purposes (PCR Insider 12/09/10).
Have topics you'd like to see covered in Pharmacogenomics Reporter? Contact the editor at tray [at] genomeweb [.] com.