Health care professionals are generally optimistic about the applications of personalized medicine in oncology and say that successful implementation of precision oncology depends on the readiness of health care practitioners to adopt the technology, along with the availability of the needed infrastructures, according to the results of a new survey.
“The findings may surprise you, as many believe that advancing science is only the first step,” said Susan Garfield, DrPH, Senior Vice President of GfK, the research organization that conducted the Novartis-sponsored survey of approximately 230 health care professionals—more than half of them oncologists, hematologists, surgeons, and gynecologists.
“Practicing oncologists need to think about using precision oncology for their patients who do not have any standard therapies available,” said Lee Schwartzberg, MD, Chief of the Division of Hematology Oncology at the University of Tennessee and Medical Director at the West Clinic in Memphis, Tennessee. “Get them into trials to give them access to drugs. Make the testing operationally useful so you can use precision oncology for your patients.”
In an interview, Schwartzberg said he has become excited about the promise of using molecular alterations to make treatment decisions: “We have tangible results in terms of more specifically targeted drugs and companion diagnostics to pick the right therapy, and an increasing awareness of what happens after resistance occurs.”
Some cancer experts, though, are less sanguine about the promise of precision oncology. “The reality is that there are a few wonderful examples of precision medicine, but they are few and far between, and they are very expensive and often overhyped,” said Leonard Saltz, MD, Professor of Medicine at Weill Cornell Medical College and Chief of the Gastrointestinal Oncology Service at Memorial Sloan-Kettering Cancer Center.
“Add to all this that ‘precision’ or ‘targeted’ medicines have some awfully nasty side effects. The idea that we would make cancer drugs similar in toxicity to other medicines used to treat other diseases is thus far just a fantasy. For most people in most common diseases, precision medicine is not a step forward.”
4 Key Areas of Focus
In the survey, the respondents identified four keys areas of focus for successful implementation of precision oncology: collaboration, professional capabilities, care processes, and systems and infrastructure. The consensus was that a new level of collaboration among oncology care professionals and across health care systems is necessary to enable effective patient care.
“Personalized medicine will place new demands on health care professionals to understand and interpret data and manage increasingly complex disease patterns and treatment strategies,” Garfield said. “Process improvements need to happen in parallel with scientific advancements to deliver personalized care. Funding, reimbursement and information technology systems must adapt to enable patient access to innovation.”
Oncologists will need to expand their knowledge base to include the increasing number of new precision medicines, Schwartzberg added. “We need to become experts in multiple diseases and understand their biology if we want to use precision oncology to the maximum. There is data emerging almost every day. Either we become more subspecialized or gain access to resources and experts to interpret the data.”
He acknowledged that it may be difficult for an oncologist to see the same volume of patients when prescribing precision medicines.
One-third of health care professionals interviewed for the survey said they believed that communication and collaboration regarding personalized medicine in oncology care currently is effective and efficient. American respondents were more skeptical: Only one-quarter of those in the U.S. said they feel that today's precision oncology is working well; in comparison, two-thirds of those in Italy said they do.
The promise of precision medicine is “the matching of effective therapies to specific cancers based on molecular drivers, and as a result maximization of benefit and a minimization of toxicities,” Saltz explained. “Everyone still harkens back to Gleevec [imatinib], a drug that is now so old that it is about to go off patent. It works very well on a rare disease, chronic myelogenous leukemia, because CML happens only when one particular molecular event happens. That is why it is so rare.”
Schwartzberg agrees that in solid tumors, it is unlikely researchers will find another success story like imatinib. “In CML, there is one alteration that doesn't change, and there are no escape pathways for the mutation, but that doesn't mean we can't find multiple ways to target solid tumors.”
There is a rationale for combination therapies. Cancer can develop through multiple abnormal pathways and at multiple points within a single pathway. Some “smart cancers” can mutate and create “escape routes” through new pathways, which can lead to drug resistance. A strategy of combination targeted therapies may block multiple targets, stopping escape routes, preventing resistance, and improving outcomes.
Human epidermal growth factor receptor 2 (HER2)-positive breast cancer is another positive example, Saltz said, and one “where we can say that a targeted approach has made a real, clinically meaningful difference.”
Schwartzberg noted that HER2 amplifications or mutations are found in a variety of cancers at a low frequency beyond breast cancer.
Another pathway under examination is the translocation of the ALK (anaplastic lymphoma receptor tyrosine kinase) gene. The ALK alteration has been associated with anaplastic large cell lymphoma, non-small cell lung cancer (NSCLC), breast cancer, and colorectal cancer. For example, NSCLC patients who have a translocation in the ALK gene or in the ROS gene respond to crizotinib, the first-generation ALK inhibitor, and there are several other drugs in development now that are even more potent for this particular alteration.
But Saltz noted recent research showing that crizotinib is not cost-effective for ALK-mutated lung cancer. “And don't forget that ALK-mutated lung cancer is only in the low single digit percentages of all lung cancers,” he continued. “Epithelial growth factor receptor-mutated lung cancer was already being treated with Tarceva, along with everyone else with lung cancer, before we finally got people to stop prescribing it for all patients, and now we give it to only about one in five patients who have an EGFR mutation,” he said.
“KRAS mutations are found in almost one quarter of lung cancer patients, but nothing targeted will work for them. Remember, before precision medicine, lung cancer was the number one cause of cancer death. We should not lose sight of the fact that it still is.”
As for colorectal cancer, “we have found that you can now use all-RAS genotyping to exclude more than half of the patients from treatment with cetuximab or panitumumab,” Saltz said. “The ‘precision’ part is just saving the money and expense of not treating more than half the people with an agent that will not help.”
Schwartzberg, however, says that he sees opportunities with precision treatments—in particular using selective BRAF and selective MAPK kinase (MEK) inhibitors now on the market. “The B-raf gene mutation is found in lung, gastrointestinal, endocrine, and colon cancers. Clearly, it is a common pathway abnormality. We now see critical oncogenes changes in terms of HER2 mutations and more downstream alterations, such as in BRAF and MEK,” he said.
Another essential part of precision oncology is diagnostics, and one third of cancer medicines in late-stage development currently use a diagnostic test to identify which patients are likely to respond.
Schwartzberg said: “I don't use tests to make initial treatment decisions other than where there is an approved drug. This is critical in NSCLC, where up to one-third of patients may have upfront mutations. If I can't use precision medicine in first-line therapy, then I want to be prepared for the next line of therapy.”
Although virtually all (98%) the respondents interviewed for the survey said they expected that personalized medicine will have a positive impact on oncology care, 70 percent agreed that there are significant challenges to the implementation. Approximately 90 percent said they expect companion diagnostics to provide at least the same or greater clinical value for oncology care in the next three to four years.
One challenge is acquiring enough tissue for appropriate testing. “Technology is better, but we still need tumor tissue for testing,” Schwartzberg said. “In some cancers, that is not available, and we may require an additional biopsy.” He said he has high hopes for a liquid biopsy approach that probes a patient's blood using circulating DNA or tumor cells to look for specific mutations. The technology is available now, but requires additional validation.
There are also issues with testing. “We need to pick the right test and work with companies who have experience with next-generation sequencing. The market has to be defined. We also need to coordinate with the laboratory and pathologist. Some tests take time—up to several weeks—but many cancer patients can't wait that long for a new therapy,” Schwartzberg said.
Then there is the matter of interpretation: “What do the tests mean? Labs are getting better at providing annotated reports to help oncologists interpret data. The results are still complex and many clinicians do not understand the significance. Using the information to make treatment decisions can be difficult for a busy practitioner who may not have the time or resources to fully interpret the reports,” he said, adding that many laboratories now provide access to a molecular geneticist.
In his hospital, “we have a clinical genomics tumor board that meets every week to review all results sent out by our physicians. We help them interpret the results and send them our group's recommendations.”
The survey respondents cited education as a necessity to achieve full implementation of personalized medicine. However, some of them also see it as an impediment to expanded use. Both reimbursement and funding are also necessary.
“Reimbursement is driven by value,” Garfield said. “There has been a relatively positive response to personalized medicine because it is a cost-effective approach to care. Payers around the world have been receptive to personalized oncology.”
“We can do broader molecular analyses now, and insurance companies are becoming better at reimbursing for them,” Schwartzberg said. “This makes it easier from both the patient and provider perspective. Panels are giving us actionable mutations, and this gives us more latitude to what we do. We can find out exactly what is happening to the patient beyond diagnostics.”
Saltz pointed out, however, that for now, the main source of information about personalized medicine comes from the burgeoning industry of expensive molecular analyses, as well as from the pharmaceutical industry.
The biggest flaw in precision medicine, Saltz said, is that “the benefits thus far tend to be very short-lived before other mutations emerge and disease control is lost. Cure is virtually unheard of. Median disease control is usually measured in months, and rarely approaches anything close to a year. As such, patients are too frequently experiencing an ‘is-that-all-there-is’ phenomenon. When they get precision medicine, all too soon they are looking into the abyss again.”
Schwartzberg agreed, but said progress is starting to accelerate: “There is a lot of hype or enthusiasm with any new modality. I was skeptical until a year ago. Anything with life and death, people want to believe. A lot of money has been spent on genomics that has not translated into many drugs yet.
“We have only done genomics for a few years, so our understanding of the range of mutations and alterations is incomplete,” he continued. “It has been only about a dozen years since the human genome was sequenced. In the last few years, multiple molecularly targeted therapies have been developed. Do they cure cancer? No, they target alterations, and provide incremental clinical benefit.”
He said he believes multiple targets are probably needed for more meaningful benefit, and cites the example of combined treatment with dabrafenib, a selective BRAF inhibitor, and trametinib, a selective MEK inhibitor, in the treatment of metastatic melanoma. “The weight of evidence is moving away from doubters. Melanoma, a disease that typically was fatal within months, now has a median progression-free survival of one year or more. That's meaningful benefit,” he said.
Schwartzberg acknowledged that there have been no inroads made in prevention strategies. “We need to understand the molecular alterations that occur in the run-up to invasive cancer. If a target is identified, we may be able to intervene before disease develops. That might, in the long run, be where precision medicine has its greatest impact.”
‘Time to Get on Board’
The time to be skeptical about precision medicine is rapidly fading, Schwartzberg maintained. “It's time to get on board. Not everyone needs to embrace precision oncology, and not every patient needs whole exome sequencing. But it is the future. More therapies will be targeted.
“After all, chemotherapy targets DNA or microtubular damage. Clearly, it targets molecular events. We didn't have the tools to understand that in the past, but we do now.”
The continuing challenge for oncologists will be to keep up with the ever-growing amounts of information and choose the right treatment. “The clinical trial system doesn't allow drugs to be tested quickly,” Schwartzberg said. “Everyone will test drugs on different patients. I strongly encourage an ‘n of 1’ data or registry approach. We will miss a valuable opportunity if we don't aggregate that data. If we do, the promise of precision oncology will be realized faster.”