I saw Mr. Jones in consultation for his newly diagnosed lung cancer. He came in for his first oncology consultation to find out what we would recommend for his treatment and what the future held for him. His grown up children and his wife accompanied him, and as is the norm nowadays, they had looked up his diagnosis on the web. As is common, this led to more concern about what was truly applicable to him.

            At most community cancer centers, you are more likely to see a general oncologist than an oncologist who specializes in only one kind of cancer.  In the last ten years, there has been an explosion of treatment options including a variety of treatments targeting immune pathways, growth receptors, genetic mutations in addition to standard chemotherapy and hormones, radiation and surgery. Depending on the extent (stage) and type of cancer, a number of these treatments are used in combination and in sequence. To make sure I am doing the right thing by my patients, I check the available guidelines. The most common guidelines that are referenced by most oncologists are the National Comprehensive Cancer Network (NCCN) guidelines.  Panels of oncologists who specialize in each disease formulate these guidelines. They meet and review the available studies and make recommendations accordingly. These determine which treatment is started first, which treatments are alternate options, which treatments are reserved for relapses and beyond.  As practicing oncologists, we depend on these recommendations to be free of commercial bias. And we expect that recommendations provide a meaningful benefit justifying toxicity and cost.

            As you can imagine, when new (and expensive) drugs make it up the usage ladder, the volume of usage increases dramatically, and nicely adds to the bottom line of the manufacturer. So there is a huge incentive to get into first line recommendations and improve the odds of usage. The experts on these committees get a lot of funding from the manufacturers, both in terms of research support as well as personal payments to be on the lecture circuits, advisory boards and other panels. I have also sat on a couple of advisory boards, which seemed more like a discussion on whether the data on the drug in question would be adequate for practicing oncologists to make changes in their practice. They were interesting discussions, but we clearly didn’t have the power to change national and international practice.

            A few years ago, a guideline committee made a recommendation that made me look at it closely. It didn’t seem to be justified by the data, and I started looking at the committee makeup and whether their funding could have influenced the recommendations. Unfortunately due to lack of available details, I couldn’t take that further, and the blog languished in my computer.  This remained an unanswered question for me.

            So I was happy to see an article by Liu et al in The Oncologist, a journal for practicing oncologists addressing this question in their April 2019 issue, They looked at NCCN guidelines, the strength of the recommendations and the authors of the recommendations in 2015.          

Financial issues: They did find that 87 % of the authors received payments, $115 million for research payments and $ 4.6 million for personal payments, including consulting fees, speaker fees, etc. 51.7% of the authors received more than $10,000 in personal payments and 78.4 % received more personal payments than research payments and 33% received personal payments only. 3 members (.6%) had ownership interests.

Strength of recommendations: These conclusions were equally bothersome, that 71.9% of recommendations were based on low-level evidence. Even 21.7 % of Category 1 recommendations were based on low-level evidence. Clearly high-level evidence does not exist in many situations and the panels had to make do with what data they had. Because of lack of clear-cut data, many recommendations were bundled into multiple options, all based on low-level evidence. There is concern about unintentional bias from industry related payments influencing the recommendations. Apparently, these authors found that although financial conflicts of interests were prevalent in the authors of NCCN guidelines there was no obvious evidence that it contributed to inclusion of systemic treatments based on low-level evidence.  However there were limitations in the information about payments available to the authors.

The continuing concern is that the guidelines have become the main basis of practice and payments in the United States. Twenty-five years ago, when I first started practicing, it was routine for doctors in the United States to be wined and dined lavishly by industry, be awarded tickets to ballgames and taken on trips. After the Sunshine Act was passed in 2009 transparency about payments from pharmaceuticals and device manufacturers to physician was required, resulting in public access to conflict of interests. There has been a change in the regulations and attitudes, recognizing that bias enters the picture when payments in cash or kind are made.

Panel members do declare “conflicts of interest”, but those may give us a false sense of transparency and objectivity. As practicing oncologists, we need to be careful about using the guidelines without also looking at the data. A spotlight needs to shine on guideline panels, as they have the power to influence practice and payments.  Ideally, guideline panels should not have members receiving personal payments of any sort.

1) http://theoncologist.alphamedpress.org/content/24/4/498.abstract

2) https://www.congress.gov/bill/111th-congress/senate-bill/301