Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications
As regulatory framework gets fine-tuned, firms realize the importance of having a biomarker strategy.
Biopharma companies are more focused on developing companion diagnostics for cancer drugs despite challenges and initial trepidations over this business model. Indeed, FDA’s recently issued draft guidance on companion tests is testament to the fact that the field is advancing and more regulatory clarity is needed.
Therapeutic firms are pursuing two strategies in developing diagnostics: through in-house work or through external partnerships with diagnostics companies. Depending on the availability of a clear-cut biomarker, the ideal is to start diagnostic development early in the clinical development cycle of a drug. The aim is to have a test ready for use in Phase III drug testing.
Firms are making the investment in companion diagnostics with the belief that it will ultimately impact the overall cost of drug development. “While there is some up-front investment, in the ideal scenario co-development of diagnostics will reduce drug development costs,” David Reese, M.D., Amgen’s vp of medical sciences, told GEN. “By allowing you to target much earlier in the drug development process patients that are likely to respond, you can demonstrate the value of the drug early on.”
Developing Dx and Rx Together
Roche’s recent advances with its metastatic melanoma candidate, vemurafenib developed through its Genentech subsidiary, and its corresponding companion diagnostic, the Cobas 4800 BRAF V600 Mutation Test developed through Roche Diagnostics, provides an example of a successful internal collaboration.
Vemurafenib is being co-developed under a 2006 license and collaboration agreement between Roche/Genentech and Plexxikon, the original drug developer. The NDA and MAA for the drug, submitted last May, covers the treatment of individuals with BRAF V600 mutation-positive metastatic melanoma.
The Cobas 4800 BRAF V600 Mutation Test, a PCR-based companion diagnostic, is designed to identify people whose tumors carry the BRAF V600 mutation. About 60% of melanomas have this mutation; melanoma cells lacking it are not inhibited by the drug, which stimulates normal BRAF and can promote tumor growth.
Commenting on the Phase III trial (BRIM3) evaluating the drug, Paul Brown, head of Roche Molecular Systems, said, “In BRIM3 our investigational test enabled rapid and accurate identification of eligible patients with metastatic melanoma.” The study met its co-primary endpoints of overall survival and progression-free survival. Risk of death was reduced by 63% and risk of disease progression by 74%.
One of the key factors in the successful, side-by-side development of vemurafenib and its companion diagnostic was that Plexxikon was in the Bay area, Walter Koch, Ph.D., vp, head of global research at Roche Molecular Diagnostics, told GEN. “Peter Hirth of Plexxikon had reached out to us in 2005, told us what Plexxikon was working on, and asked if we could develop a test.
“Our diagnostics agreement predated that of Roche Pharma, and having that diagnostic relationship in place played a key role in our pharmaceutical colleagues getting involved in it. Once Roche had decided to in-license and develop the drug, a lot of planning had started in 2006 to coordinate the development of the drug and the test and to have the test available for the clinical trials.”
Expected Challenges
Commenting on hurdles with companion diagnostic development for vemurafenib, Dr. Koch said the first challenge was a “generic one. Drug development and diagnostic development are fundamentally different, and aligning the processes so that both the therapeutic and the test can be filed with the FDA simultaneously works better if you have both teams at the same table together.
“This was greatly facilitated in our case because we were all the same company.” While he believes it can be done with external partners, Dr. Koch noted that it would be more complicated.
The second challenge he mentioned was linking hardware, software, and test reagents. “We were working with a different generation platform in 2005, then had to change the platform during the development process. We moved to a newer generation instrument, and so had to revalidate, re-tune, and re-optimize the reagents so they worked with the new instrument and software.”
Despite inherent challenges, Dr. Koch said that “in Roche and in most pharma companies, personalized healthcare is the way we are moving forward. We believe we will greatly improve the efficacy of therapeutics when we can identify appropriate patients. We are seeing early success rates that suggest that companion diagnostics are a far better way to improve therapeutics and success rates of clinical trials.”
Having Biomarker Data Early
Vemurafenib stands apart from other cancer drugs with companion tests because knowledge of specific mutations affecting drug response was available early in the drug’s development cycle. Dr. Koch explained that while there are multiple examples of cancer drugs with specific biomarker tests that characterize the diseases, knowledge of specific mutations affecting drug response became available after the drugs were developed. “This substantially changes the market,” Dr. Koch said.
Drugs approved along with a companion diagnostic include Eli Lilly’s Erbitux and Amgen’s Vectibix, both anti-epidermal growth factor receptor (EGFR) antibodies. The original companion diagnostic, EGFR pharmDx, which detected EGRF status in patients, was developed by partnering with DAKO.
Evidence that KRAS mutation status would affect how subjects responded to anti-EGFR treatments became apparent only after clinical trials that formed the basis of NDA and MAA submissions were completed. Amgen and Imclone analyzed the existing information and found that about 43% of patients, those with KRAS mutations, did not respond to the EGRF inhibitors.
“Therefore, if you had a diagnostic test for the KRAS mutations, you would have 40% of patients who wouldn’t get the therapy but who would have been treated prior to that knowledge.”
But, Dr. Koch said, many drugs are evolving along the lines of BRAF, “where an activating mutation in an oncogene makes sense to target specifically or knowledge of a pathway downstream portends failure for the drug and suggests other development strategies.”
Amgen has bolstered work on identifying biomarkers that will allow the company to predict which patients will or won’t respond to a given drug. Generally the company identifies a diagnostic company with appropriate expertise to partner with depending on the type of test needed.
“We believe it’s an essential part of the drug development process,” Dr. Reese told GEN. “Every product we put into humans has a biomarker team formed typically one to two years before the first patient is dosed with a drug.” He explained that the teams have two mandates. “One is to potentially develop pharmacodynamic biomarkers that tell you whether the drug is doing what you think it should be doing. For example, if you are targeting a receptor in cancer, is your drug getting in at an appropriate level and shutting the receptor off?
“The second mandate of that team is to identify those patients likely to respond or not respond to the therapy. Our goal is to get the right drug to the right patient as early in the process as possible, starting that effort even with preclinical work.”
Dr. Reese explained that what potentially differentiates Amgen from companies developing companion diagnostics for their drug candidates is “the scale of the effort and the formation of cross-functional teams with the sole mandate of biomarker development.”
Integrating Companion Dx Development
“Patient selection can take many forms, depending on the clinical scenario and really has to be customized to each setting,” Dr. Reese remarked. “There is no one size fits all development process. We have the philosophy that the science and the biology drive what we do and shape what we do in the clinic.”
Companion diagnostics offer the perfect means of fulfilling the promise of personalized medicine. They have, however, been viewed as disruptors of the biopharmaceutical market, as the tests could restrict the size of treatable populations.
Increasingly, however, pharma and biotech companies are realizing that they cannot ignore companion diagnostics. With FDA’s draft guidance on this sector suggesting that drugs and their corresponding tests must be approved simultaneously, investment in either internal strategies or external partnerships are ramping up.
Patricia F. Dimond, Ph.D. ([email protected]), is a principal at BioInsight Consulting.