Companion Diagnostics: The Future of Personalized Medicine

The future of healthcare lies in the development of personalized medicine, in which the combination of diagnostic testing and treatment therapies allows physicians to design a specific course of treatment based on a patient’s genetic makeup.

These diagnostic tests, called companion diagnostics (CDx), are still early in their inception, but they are becoming more widely developed and utilized to offer healthcare providers the opportunity to identify the exact treatments that may have the greatest impact on a patient’s condition, including specific drug dosages. In fact, worldwide acceptance of the ability to identify predictive biomarkers and subsequent CDx testing in the development of patient therapy has grown to be recognized as the next step forward in patient care.¹ With CDx testing, a patient’s therapeutic interventions can be decided upon earlier because other therapies can be ruled out. In addition, treatment can provide maximum benefit, and side effects can be minimized.

CDx testing was first introduced in 1998 with the HercepTest, a diagnostic tool developed by DakoCytomation that identified patients for whom the cancer drug Herceptin (trastuzumab) would be effective. Since then, CDx tools are proving to be effective for a more patient-tailored approach to care.² The Personalized Medicine Coalition reports a sharp rise in the number of personalized medicine drugs, treatments and diagnostic products available to consumers just in the last year.³ Not only are new drugs becoming available, but more and more research by both pharma companies and drug companies are identifying the populations they may best serve.

“We are becoming more and more successful at identifying biomarkers,” says Amy Miller, vice president of public policy at the Personalized Medicine Coalition, specifically in the area of cancer research. This is significant because, as Marc Boutin, executive vice president of the National Health Council, explains, it is not uncommon for medicines to work only 60 percent to 65 percent of the time, and in the area of oncology, that percentage drops to about 10 percent. Right now and in the near future, oncology is the one area science is focusing on in the development of CDx testing.

There were two significant U.S. Food and Drug Administration (FDA) CDx approvals in 2012. One was Pfizer’s Xalkori (crizotinib), which was approved to treat patients with late-stage lung cancer who have an abnormal anaplastic lymphoma kinase (ALK) gene due to a gene rearrangement, and which comes with a CDx test to screen potential patients for the ALK mutation. Another is Daiichi Sankyo’s and Roche’s Zelboraf (vemurafenib), which was approved for the BRAF V600E-mutation positive melanoma drug in combination with a CDx test to screen patients for the mutation.³

Some argue that the addition of CDx testing will limit the production of certain drugs because the subset for whom they will be effective is limited. However, others see the promise of drugs that did not initially receive FDA approval realizing a new life because of a new group of people for whom they may be effective. Therefore, research and development dollars spent on those drugs may see a return after all, even though the market to whom drug makers can sell may be limited.

CDx: Easing the Way Toward Effective Treatments

There is no question that the era of personalized medicine has arrived. In 2006, only 13 tests were available to determine a patient’s susceptibility to a particular disease or responsiveness to a particular treatment. Today, according to the Personalized Medicine Coalition, there are more than 72 FDA-approved therapies — a fivefold increase.⁴ “Not all of these drugs came out new,” explains Miller, “but drug companies ask for label changes based on research, or sometimes the FDA will change labels based on research.” Though past success in the personalized medicine field has been debatable, industry observers say the science, regulation and business models are coming together in favor of CDx’s promise.⁵

Today, patient therapy is largely based on demographics and both the doctor’s and patient’s hope that the therapy will work. But, because a given medicine often works in only a small percentage of the population, CDx can make an enormous difference in “getting the right treatment to the right patient at the right time.” The challenge is to change physicians’ behavior so that they test, when a test is available, for biomarkers prior to determining a treatment plan.

These biomarkers are the enzymes responsible for drug metabolism and the proteins that determine the cellular response to drugs, which can vary in expression depending on a patient’s genetic makeup. Genetic tests can determine whether a patient carries these variations, which will allow physicians to individualize drug therapy. So, rather than basing dosing on height and weight, genetic testing will tell physicians how much, how well and how long it will take for a body to process medicine. This is known as pharmacogenomics, which gets the right dose of the right drug to the right patient at the right time, and it can save lives. 6

Take, for example, rilotumumab (AMG 102), which in early studies did not fare well and, in turn, did not warrant further study for drugmaker Amgen. When the biomarker data came out for rilotumumab, the scales were tipped, and the drug is now a shining example of how genetic testing can impact the understanding of how the drugs work and for whom they’ll work best.

Rilotumumab when used with chemotherapy nearly doubles the survival rate for those with high levels of the protein c-Met, which can bind with hepatocyte growth factor and create cancer cells. Rilotumumab blocks c-Met and hepatocyte from binding and increases the survival rate for those with gastric cancers to 11.1 months, compared with 5.7 months for patients taking a placebo. The American Cancer Society estimates there will be more than 21,000 new cases of gastric cancers this year and, thanks to advancements in personalized medicine, rilotumumab may play a large role in patients’ survival.⁴

Ten percent of marketed drugs suggest genetic testing for optimal treatment.⁷ Among the most common of the several genes responsible for drug metabolism are cytochrome P450 (CYP450) genes, more than 50 encoding enzymes that control the metabolism of more than 70 percent of prescription drugs. Those who carry this gene often do not metabolize drugs at the same rate as most, leading to poor response or adverse reactions.

For example, the conversion of the drug codeine to morphine is mediated by the CYP450 enzyme CYP2D6. In some, metabolism is too slow and ineffective.In others, it is too rapid with subsequent risks of opioid toxicity. The FDA Public Health Advisory warns that the only way to assess the metabolism response prior to the administration of codeine is with a genetic test.⁶

Another example of how genetic variations affect drug metabolism in those patients whose genes encode drug-metabolizing proteins (or targets of the drugs themselves in such a way that certain drugs become ineffective) is clopidogrel (Plavix), a platelet inhibitor in patients with cardiovascular disease. It is estimated that 2 percent to 20 percent of patients carry mutations to the CYP450 enzyme CYP2C19, which makes them poor metabolizers with a reduced capacity to convert clopidogrel into its active metabolite. This results in not only a reduced anti-platelet effect, but it also makes patients more likely to have an ischemic event following therapy.⁶

Warfarin (Coumadin), an anticoagulant, is metabolized by the CYP2C9 enzyme, and its anticoagulant effect is mediated by the enzyme VKORC1. A combination of slow warfarin metabolism caused by CYP2C9 gene variations and reduced coagulation caused by VKORC1 gene variations increases the risk of bleeding during warfarin therapy. Warfarin product labeling states: “The patient’s CYP2C9 and VKORC1 genotype information, when available, can assist in selection of the starting dose.”⁶ It is estimated that 17,000 strokes could be prevented each year if a genetic test were given prior to the administration of warfarin to determine proper dosage.⁷

Now, CDx tests are being created alongside drugs, with partnerships forming between drugmakers and those in diagnostics so the drugs and tests can be approved by regulators at the same time. It’s harder to get drugs approved through the FDA if there isn’t a clear population they target, so this relationship means drugs could potentially be approved more quickly. According to the Personalized Medicine Coalition, 30 percent of treatments in late-stage clinical development and 50 percent of treatments in early-stage clinical development rely on biomarker data.⁷ Matching the right diagnostic and drug partner can make for a smoother transition from research to marketplace, and these relationships are thought by many in the industry to be the key to advancing product development.⁸

“This is an area of a lot of activity at the moment,” says Hakan Sakul, PhD, executive director and head of diagnostics, Worldwide R&D, for Pfizer Inc. “New applications are pairing drugs and diagnostics, but not everyone is used to seeing diagnostic testing on a drug label. The pharmaceutical industry is learning more about CDx, and how sharp that learning curve is depends on whether a pharma company has previous CDx experience.”

Common CDx Tests

Collaboration is under way as drug developers partner with diagnostic testing manufacturers to address the issue of matching a drug with the appropriate patient. Drug developers want assurances that their product is going to reach a certain audience and be successful in its treatment. Likewise, diagnostic testing manufacturers can provide the means to that assurance and potentially help the drug to reach FDA approval more quickly by proving effectiveness. The relationship can be symbiotic for all parties involved, from researchers to patients. “There is constant interaction between diagnostic companies and pharma,” says Dr. Sakul. “They [the diagnostic developers] come to us [the pharma industry] because they want to share new advances, new technology and new products. When the time comes to partner, we usually have a good idea of which companies to contact for potential CDx applications. This was the case for us with Xalkori. When we knew a diagnostic test would be required for FDA approval, we knew where to turn.”

Xalkori, Pfizer’s lung cancer drug, is effective only for a small percentage of patients whose non-small-cell lung cancer (NSCLC) has a rearrangement of the ALK gene on the 2p23 chromosome. About 85 percent of lung cancer patients have NSCLC, which corresponds to a very low survival rate. The drug can reduce tumor size in 57 percent of those patients and stop progression in 87 percent.⁴ Partnered with Abbott’s Vysis ALK Break Apart FISH Probe Kit, physicians can test for ALK gene-positive patients for whom Xalkori is appropriate.⁹

The biotechnology firm Qiagen has 15 signed agreements with drug developers for its molecular diagnostic testing business, including Pfizer, which is partnering with Qiagen to develop diagnostic testing for a potential lung cancer treatment.⁴

Abbott and Merck are collaborating on a CDx for an investigational cancer therapy using Abbott’s proprietary FISH (fluorescence in situ hybridization) technology to design a test for TP53 gene deletions in cancer patients. 9 Abbott also is working with GlaxoSmithKline (GSK) to develop polymerase chain reaction (PCR) tests to screen non-small-cell lung cancer and melanoma tumors for expression of the MAGE-A3 antigen and PRAME antigen in support of GSK’s cancer immunotherapy research program.⁹

Already on the market are Roche Holding AG’s Herceptin (trastuzumab) for the treatment of breast cancer (one of the first cancer medicines aimed at patients who have the genetic abnormality HER2) and a CDx called the PathVysion HER-2 DNA Probe Kit developed by Abbott that tests for the mutation.⁹

The Roche Group also has received expedited review and approval for the melanoma drug Zelboraf (vemurafenib), which targets patients with metastatic or unresectable melanoma whose tumors express a gene mutation called BRAF V600E. Its CDx, the cobas 4800 BRAF V600 Mutation Test, manufactured by Roche Molecular Systems, also was approved ahead of schedule. This BRAF mutation occurs in about half of late-stage melanomas, and Zelboraf can inhibit the function of the V600E-mutated BRAF protein. Zelboraf is marketed by south San Francisco-based Genentech, a member of the Roche Group. 10

The FDA has created a personalized medicine program that is addressing policy and approval issues of therapeutics and diagnostics. Phases I and II have been completed, and the comment period for draft guidance on novel diagnostics ended in September 2011. Comments are now being analyzed to complete the final guidance regulations. “The FDA is quite serious about pairing drugs and diagnostics and wants them to be filed [for approval] together,” explains Dr. Sakul. The FDA has made it very clear that it will require that “highest hurdle.”

Various departments within the FDA, including staff trained in pharmacogenomics and biomarker expertise, will work in tandem to ensure regulatory oversight of therapeutics and diagnostics to better enable approved products to get to market safely and efficiently. An expedited approval process for “breakthrough therapies,” which can shorten the regulatory approval process by 40 percent to 50 percent, also has been approved with the recent passage of FDA user fee authorization legislation.¹¹

Payment for CDx

CDx can save insurance companies, physicians and patients hundreds of millions of dollars annually with the reduction or elimination of both ineffective treatments and incorrect dosages. However, financially speaking, the long-term cost savings of the tests may not always be immediately evident based on the costs of the tests themselves, which in some cases can be quite high.

For example, ineffective treatments costing thousands to tens of thousands of dollars may be avoided with a simple genetic test. However, if the genetic test itself also is expensive or there is another way to test for the necessity of treatment, insurance companies may be reluctant to pay those upfront costs without valid cost-benefit and cost-comparison information. Or, insurance companies may decide all patients should be tested if there is an available diagnostic to weed out a small few for whom a drug could be unsafe, as in the case of warfarin, or they may decide it is more cost effective to just remove patients from the drug should adverse effects be observed.

That being said, there is no standard method of payment for CDx testing, and reimbursement can range from just a few dollars to tens of thousands of dollars, with rates set in most cases by the Centers for Medicare and Medicaid Services (CMS) and Current Procedural Terminology (CPT) codes that provide a payment standardization for all payers.

At present, reimbursement is determined on a case-by-case basis based on whether the insurance companies deem the test appropriate for the course of treatment. One exception appears to be when a drug and its CDx come to market together. This inconsistency in reimbursement has led to confusion in coverage decisions and, it is feared, could hinder the development of the next generation of products, reducing access and impeding the progress of personalized medicine.

While the tests are of high value to the patient and high value to the system, as the National Health Council’s Boutin explains, they are harder to value for payers. Not only are clinical trials conducted for FDA approval of the drug, but payers as well want to see a comparative advantage analysis to make sure the drug will be of value.

It is important before a product goes to market that the developer provide a clear demonstration as to the cost-benefit of its tests through comparative effectiveness research. For instance, it is estimated that there would be a 34 percent reduction in chemotherapy use if women were given a genetic test prior to treatment for breast cancer⁷ — a huge cost savings financially, emotionally and physically. Even with this evidence, insurance companies do not have a clear benchmark indicating which specific improvement would warrant what payment. There is clearly confusion in the marketplace, and this confusion will remain for the next few years.¹²

However, this confusion could be remedied with the passage of a bill called the MODDERN Cures Act, introduced by U.S. Rep. Leonard Lance of New Jersey and the National Health Council. The MODDERN Cures Act aims to reinvent the drug and diagnostic regulatory framework in part by extending a drug’s pharmaceutical patent so that it doesn’t expire before or soon after the drug is brought to market, providing patent protection from generic competition. The legislation also encourages the co-development of diagnostics and drug therapies. As of this writing, there are 35 House members on board, and it will be introduced into the Senate soon.

“This [the MODDERN Cures Act] could double or triple the investment in companion diagnostics and changes the business model,” says Boutin. It will “change the insurance conversation with beneficiaries” as well. This is because as CDx are more predictive and more widely used, patients with the same disease will no longer be treated with the same medications, but ones tailored to their genetic makeup.

The Future of CDx with Regard to Personalized Care

While it appears the huge breakthrough in CDx is turning the tides of medicine, it is important to remember that science is still early in its journey and there is still much to be learned. New technologies will be developed to better understand disease and to better sort through the vast array of information we are gathering now and will gather in the future as science and technology progress. It is an exciting time for medicine and for patients and an exciting time to be on the ground floor.

“The future of diagnostic testing is in part being driven by prevailing science guiding where to focus efforts,” explains Dr. Sakul. “Companion diagnostics can help to provide better opportunities for patients.” Right now, the predominant focus is on cancer therapies. However, as science turns towards other diseases, drug and diagnostic companies will jump quickly to develop protocols. If a patient is diagnosed with cancer, learning which drug will be more likely to work will better benefit not only the patient but healthcare overall.