Oncology Biomarker Diagnostics: Where We Are, Where We Need To Be

Despite the promise of biomarkers for use in diagnosing cancer, there have been few commercialized biomarker diagnostics.

Margaret Piper, PhD, MPH, director, genomics resources at the Blue Cross Blue Shield Association Technology Evaluation Center, points out that “oncology pathways are the result of accumulating mutations that disrupt normal pathways and cause changes in growth and metastases. This creates challenges in biomarker identification and validation.” So, Piper says, it can be very difficult to tell which genetic mutations are most important — that is, which are more indicative of aggressiveness or impending metastases rather than “passenger” mutations. The individualized nature of different cancers in different people adds more complexity in terms of the science.

Then, there are intellectual property (IP) and commercialization issues, which makes developing diagnostics and getting them to market even more complicated.

Why biomarker diagnostics?

In a February speech before the Personalized Medicine Coalition, U.S. Food and Drug Administration Commissioner Margaret A. Hamburg, MD, underscored the need for diagnostics development. She emphasized that when patients are battling cancer, “...they don’t have time for trial and error; prescribing an ineffective course of treatment wastes precious time in the fight against a potentially fatal disease progression, and the toxicity from exposure to an effective therapy can undermine their efforts.”1

The cost of new cancer drugs is generally over $5,000 per cycle. Combination therapy is routine, and supportive care agents to reduce toxicity or improve quality of life often are needed. In its 2009 Drug Trend Report,2 Express Scripts forecasts a 24 to 25 percent rise in the cost of oncology agents from 2010 to 2012, and notes that cancer is now a chronic condition (Figure). The Medical News3 estimates that as of January 2010, there were 725 cancer drugs in development.

A strong proponent of oncology biomarker diagnostics, Bruce Quinn, MD, PhD, formerly contractor medical director for the California Medicare Part B program and now a senior health policy specialist with the law firm Foley Hoag in Boston, notes that the annual cost of one of the newer chemotherapy agents can exceed $50,000.

Massimo Cristofanilli, MD, FACP, chairman of medical oncology, at Fox Chase Center, in Philadelphia, states that although several thousand dollars for a diagnostic test may seem expensive, other diagnostics — such as positron emission tomography and computerized tomography, which are used to assess response to chemotherapy — may cost as much or more.

Greg Weinhoff, MD, managing partner of CHL Partners, in Stamford, Conn., a life sciences venture capital firm, confirms that investor attention is focused on molecular diagnostics that help clinicians choose between costly drug alternatives or avoid the use of toxic drugs or unnecessary interventions.

Steven Tuetsch, MD, MPH, chief science officer in the Los Angeles County Public Health Department, notes that the substantial monies and focus directed to comparative effectiveness research may well highlight the value of molecular diagnostics. These tests, Tuetsch says, help oncologists compare oncology drugs with alternatives, e.g., a drug’s relative clinical utility. Tuetsch also notes that molecular diagnostics can help assess treatment cost-effectiveness, which, with the large number of oncology drugs in development and their higher cost, will be important from a societal cost perspective.

Biomarkers also are needed for the development of new oncology agents. The FDA’s Center for Devices and Radiological Health is expected to publish a revised guidance on companion diagnostics later this year. Nancy Bradish Myers, president of Catalyst Healthcare Consulting, in Washington, says this will help innovator companies develop a diagnostic development process that is in tandem with the FDA’s approval process. In regard to the oncology biomarker diagnostics that may be commercialized in the next several years, it will be interesting to see how many come out of the FDA companion diagnostics process and are co-marketed with oncology drugs versus those that are commercialized through an alternative route.

The I-SPY2 trial

The Biomarkers Consortium, led by the Foundation for the National Institutes of Health, is a public/private partnership of the FDA, National Institutes of Health (NIH), and leading pharmaceutical companies. Last March, the consortium launched the I-SPY2 trial, which uses biomarkers from women with newly diagnosed invasive breast cancer to assist in screening new drugs. The first agents to be tested include two from Amgen (an APO/TRAIL inhibitor and an angiogenesis inhibitor), two from Pfizer (an IGFR inhibitor and a Pan ErbB inhibitor), and one from Abbott (a PARP inhibitor). It is notable that these agents represent five different therapeutic classes with different mechanisms of action. The frequent use of combination therapy for cancer may well enable the I-SPY2 trial to provide even more data about treating aggressive breast cancers. A unique feature of this trial is that it is adaptive — data from one set of patients will be used to help guide decisions about which treatments are likely to be more clinically beneficial for another, and later, set of patients. At the same time, the trial is not designed for prospective biomarker validation. The I-SPY2 trial is expected to be beneficial in selecting oncology agents from both efficacy and safety/quality of life perspectives.

The IP issue

Quinn has raised a concern about the availability of oncology biomarker diagnostics for agents studied in National Cancer Institute (NCI) clinical trials.

The NCI has proposed a revision to its IP agreements with certain funding recipients (investigators and cooperative groups) called the Cancer Therapy Evaluation Program (CTEP) Intellectual Property Option.4 This option says that biotechnology and pharmaceutical manufacturers providing agents for use in a CTEP-supported trial will have substantial control over any worldwide commercialization of any “inventions” identified during the CTEP trial. The proposed revision would apply to inventions that “are conceived or first actually reduced to practice” pursuant to NCI CTEP-funded clinical or nonclinical studies utilizing the agent(s). It also applies to inventions pursuant to NCI CTEP studies utilizing clinical data or specimens from patients treated with the agents (including specimens obtained from NCI CTEP-Funded Tissue Banks).4 Biologic manufacturers would receive a “paid-up” worldwide license to these inventions for research purposes as well as a time-limited first option to negotiate a royalty-bearing worldwide license for commercial purposes.

One of the concerns about this proposed NCI policy is that it may limit the development and availability of cancer biomarker diagnostics and reduce the participation of investigators and oncology cooperative groups in NCI-funded clinical trials, which are substantially funded with public monies.

Essential for personalized medicine

Several people interviewed for this article noted that they expect a relatively small number of valuable oncology biomarker diagnostics to become commercially available in the next several years.

Clinical trials, such as I-SPY2, will require some time to develop and publish the findings. As more cancer biomarkers are discovered, however, molecular diagnostics will be even more necessary. Cristofanilli estimates that in 2 to 3 years, there will be more biomarkers for breast, head and neck, and prostate cancers, which will have a large impact on community oncologists’ practices. Non-small cell lung cancer is another condition where a number of biomarkers may be identified.

In June, the American Society of Clinical Oncologists (ASCO), which has been very active in the area of biomarkers, published its third guideline on germ cell tumors.5 One benefit of treatment guidelines from professional societies such as ASCO or the National Comprehensive Cancer Network is that payers consider these guidelines to be very important in making coverage decisions, particularly in a complex, rapidly evolving, and individualized therapeutic area such as cancer where clinical evidence to support a new diagnostic is lacking. Payers interviewed for this article expressed frustration with the very limited data for diagnostics, particularly when their price is high, which may lead some payers to not cover them or to delay coverage until better evidence is available.

Another factor is that the FDA regulates in vitro diagnostics as devices. Device manufacturers primarily use the 510[k] process to gain marketing approval, because it is fast, streamlined, and requires only that the device show equivalence to current devices. Very few devices are approved through the more rigorous, time-consuming, and costly premarketing approval (PMA) process, which has the advantage of requiring more evidence of safety and effectiveness but creates a higher entry barrier for competing products. Myers, with the Catalyst Health Consortium, says that the FDA may be reviewing its regulation of diagnostics via the 510[k] process, and Tuetsch, in Los Angeles, says the FDA also is taking a “harder look” at genomic tests.

Another issue from a payer coverage-and-reimbursement perspective s how oncology biomarker diagnostics affect clinical practice. As one medical director noted, “We need clinical utility and validity, sensitivity and specificity; identifying nonresponders is the return on investment (ROI).” At the Medicare Coverage Advisory Committee meeting this past January, one committee member noted that, “Raising the bar [for oncology biomarkers] is almost equal to a new drug,” while another asked, “Is it likely to be of value in community practice?” The KRAS diagnostic test, for example, by identifying nonresponders to panitu-mumab (Vectibix) and cetuximab (Erbitux) saves costs and is also inexpensive and widely distributed. Accordingly, payer coverage is very favorable.

Sarah Collins, MBA

Manufacturer concerns

From a diagnostics manufacturer’s perspective, there are several significant issues. Development of clinical data to substantiate test value is likely to become increasingly important not only for regulatory approval, but also for clinical development strategizing. Weinhoff, at CHLPartners, comments that the “venture and public funding bar has gotten higher,” which also may influence which oncology biomarker diagnostic gets to market.

Additionally, MCOs typically contract with a national laboratory, like Quest Diagnostics or Laboratory Corporation of America, for most laboratory tests, which are low cost and commoditized. Regional laboratories and hospitals account for much of the remaining market. Accordingly, a biomarker diagnostic manufacturer has to determine its distribution strategy — limited distribution may restrict market volume and access, but a deal with a major laboratory may substantially limit potential profits.

Another influential issue is the recent court ruling that seven Myriad Genetics patents are invalid, a decision that the company has appealed to the Court of Appeals for the Federal Circuit. W. Edward Ramage, JD, chairman of Memphis-based Baker, Donelson, Bearman, Caldwell & Berkowitz’s Intellectual Property Group, expects a decision in about a year and believes that, ultimately, the Supreme Court may address the issue of the patentability of genes. This issue will have a substantial impact on the oncology biomarker diagnostic industry, for if genes are found to not be patentable, this could lead to the availability of lower-cost tests that cover more genes.

Moving ahead

Based on my interviews with the persons mentioned in this article, as well as my own research, here are some important steps that developers of oncology biomarker diagnostics should consider:

• Conduct a realistic assessment of whether the diagnostic has a compelling clinical value proposition from the perspective of patients, oncologists, and payers.

• Develop strong clinical data that will be valuable to both oncologists and payers and communicate that value. Spending to develop quality data is likely to facilitate adoption, uptake, and payment.

• Assess the economic value of the oncology biomarker diagnostic to payers, not only as an ROI (typically when a diagnostic guides therapy selection), but also as to how strong the diagnostics data are for the test to lead to a change in clinical decision making.

• Implement a value-based pricing strategy. Take into account both the clinical value (notably what clinical decisions the test informs) and the economic value.

Also, keep in mind that the use of oncology biomarkers to enhance efficacy and improve quality of life will be an important component of personalized cancer treatment and management.