Oncology Drug Labels Need Patient Outcome Data: Basch Explains

ethan basch.jpg

Despite various efforts by the FDA to include patient-reported outcome data in product labeling, this sort of information is lacking, especially for cancer treatments. For instance, in 2011, there were 15 hematology-oncology drugs approved by the FDA, but only one had symptom information in the label. And only about 24 percent of prescription drugs, in general, that were approved between 2002 and 2006 had listed the effects on patient symptoms or functioning, according to Value Health. Ethan Basch, an oncologist who heads the Cancer Outcomes Research Program at the University of North Carolina Schools of Medicine and Public Health, argued in a recent piece in The New England Journal of Medicine that both drugmakers and FDA should push harder to include this information in labeling. We spoke with him about why it is important to know how patients feel before, during and after treatments…
Pharmalot: So why don’t the labels have this kind of information?
Basch: There are cultural and historical reasons and perceived logistical barriers that have prevented people from keeping this information in when implementing drug development programs and planning studies to meet regulatory expectations. Historically, and until very recently, oncology was much more focused on very small improvements in survival. But the context has changed somewhat substantially now in that we have many people who are living much longer after cancer treatment. And people are also undergoing treatment for much longer periods of time with multiple lines of therapy.
So it’s become more important how people feel before, during and after treatment. But our sensitivity hasn’t grown with that. Instead, we’re looking at serologic biomarkers as well as survival endpoints, such as progression free survival, and radiographic measures, such as tumor response, which many feel is not always clinically meaningful. We’ve become very focused on those as endpoints of cancer trials and evaluations of efficacy of products.
What we’re not focused on is how these products make people feel. Whether it’s an assumption that someone may get better or a symptom related to toxicity from a treatment. And unfortunately, it’s very uneven the way we measure these things. And we’re not really sure how the products make people feel at the end of the process, and it almost never winds up in drug labeling.
Pharmalot: You say that it’s important how people feel. How might the labeling make a difference in terms of drug development and our understanding of patient reactions?
Basch: If the regulator requires the information to be rigorously collected in order to understand the characteristics of a product or to get information in the label, then the sponsor will adhere to standards. That means it will actually believe the data being collected – it will be more unbiased, there will be less missing data, the kinds of questions asked will be more thoughtful. The methodology will be as thoughtful for other regulatory grade endpoints. So if it’s part of the label and part of the whole program, it’ll be done better and we’ll believe information more.
Second, if it’s in the portion of labeling that can be cited for marketing, they can use that information in the US to market the product and to explain benefits. That’s helpful to providers and patients for understanding (the product). If it’s not in the label and not marketed, often time people don’t find out about it.. When I sit down with patients, I actually don’t know often times how (a medication) will impact the way people feel. There isn’t a systematic approach to providing information in ways that people can feel comfortable.
Pharmalot: Why has there been a difference between oncology drugs and other drugs?
Basch: I think divisions within FDA have developed differently. In rheumatology, what’s the most important outcome? Pain functionality and mobility, for instance, and I think the information is most appropriately collected from patients. Similarly with analgesics - with a headache, the best way to find out (about the effect of a medicine) is to ask people about how they feel. GI is more similar to oncology – you may have objective measures such as motility, but you can also ask patients about their swallowing…So this is evolving over time, but the culture of the disease and the ways people think about disease and the FDA divisions thinking about endpoints.
But I think it’s related to oncology – our outcomes were not good. Sadly, people died and we didn’t have great survival (results) and so our thinking is still caught up in that context. We’re living in a new era, though, and people are living longer and how they feel affects more people than in the past. Invariably, I think there’s been increasing focus on how people feel.
Pharmalot: What’s a good example of drug in which labeling has included this sort of information?
Basch: Go back to 1996, when mitoxantrone was approved in prostate cancer based on its ability to palliate pain. There’s a huge span between then and now, when another drug – Zytiga – most recently got a labeling claim for delay in the development of pain. In both cases, understanding the pain experience was very important. Men can experience very substantial pain when it metastasizes to bone. Most will experience this and to be able to include this kind of information in the labeling is a very positive development. But it’s challenging to measure this kind of information – the patients are ill and in end-stage disease and they’re being asked to provide this kind of information, but it’s feasible. With Zytiga, pain was measured in large multinational, Phase III trials… and they did so with excellent compliance. To me, it gives their product a competitive advantage.
Pharmalot: Can you explain that point?
Basch: What’s starting to be seen is that some insurance carriers are starting to cover Zytiga instead of Xtandi, which is a Medivation drug. Both were approved with the same overall survival benefit but only Zytiga had pain in the labeling. Pain data can be informative. Quite separately, we know payers are very interested in data on symptoms. We know that people who have more symptoms need more services, which can be expensive. So having this information can be helpful. Not having the information can lead to an incomplete picture. I do think industry is increasingly thinking about collecting this information, not necessarily for regulators, but for payers. The intermediary stop is regulatory review and so it makes sense to get in the label.
Pharmalot: What do you think must be done before the FDA requires this uniformly?
Basch: I think industry responds to payers and to regulators, of course. On the regulatory side, I don’t think you can say require. I think that when the FDA meets with sponsors, if the FDA makes it clear that in order to understand the characteristic of a product, there must be an understanding of the impact on way people and function. If FDA furthers makes it clear, information must be collected I think industry will respond. Unfortunately, it usually doesn’t even come up in meetings between FDA and sponsors or it comes up because sponsors ask a question about it.
To me, the FDA needs to be proactive. They must come to the conclusion that we have an incomplete understanding of products and the only way to understand is to ask people. Until the FDA expresses that in their direct meeting with industry, industry will not necessarily respond.
Pharmalot: And how does one make that happen?
Basch: I think it’s at the leadership level. A message has to be conveyed that when individual reviewers are meeting with sponsors, there must be a discussion about how the patient experience is going to be measured. Until they do, I don’t think it’ll be at forefront of peoples’ minds.
Some of the PDUFA V funds were earmarked for expanding FDA expertise around patient reported outcomes and it’s been caught up in administrative red tape. I think the FDA has expressed interest in expanding its expertise, but their hands are a bit tied right now. But this shouldn’t stop any other effort from proceeding.