NCCN USES WEAK EVIDENCE FOR THEIR GUIDELINES

The National Comprehensive Care Network (NCCN) has a large influence on the use of cancer drugs in the United States. Its recommendation of a drug for a particular indication is an important endorsement because it often ensures reimbursement from government and private payers. 

The sway of the NCCN, which is located in Fort Washington, Pennsylvania, is partly the result of 25-year-old federal legislation that allows use of expert-generated compendiums to help determine coverage and reimbursement for unapproved use of drugs in cancer care. The NCCN clinical guidelines were recognized as one such compendiums in 2008.

Now, for the first time, a study compares NCCN recommendations to US Food and Drug Administration (FDA) approvals and concludes that the NCCN commonly recommends indications beyond what the FDA approves and largely does so using "weak" evidence.

The study was published online March 7 in BMJ.

The investigators looked at a set of 47 new anticancer drugs approved by the FDA between 2011 and 2015. Overall, the FDA granted these drugs 69 indications.

For the 47 same drugs, the NCCN recommended the same 69 indications — plus 44 other indications, report senior author, Vinay Prasad, MD, from the Oregon Health & Sciences University in Portland, and colleagues.

Prasad and his team scrutinized those additional 44 indications that were recommended by the NCCN (and not approved by the FDA) and found that only 10 (23%) were based on evidence from randomized controlled trials, which is considered "strong" evidence. The other, weaker evidence included expert opinion, case reports and series, phase 1 data, or phase 2 studies without randomization or with fewer than 50 patients.

During the same period, 58% of all FDA cancer drug approvals were based on randomized trials (ie, strong evidence), the authors point out.

The authors also suggest that there is a need to examine and possibly change the process whereby compendiums such as the NCCN guidelines act as surrogate authorities for Centers for Medicare & Medicaid Services (CMS) and deeply influence coverage decisions by payers.

"We have a system that needs an audit," said Prasad in comments to Medscape Medical News.

The study authors also highlight 2016 research on NCCN guideline panel ties to industry. As reported at the time by Medscape Medical News, that study found that 86% of NCCN panel members, who are from 27 member US cancer centers, had financial ties to industry, with 84% receiving personal payments and 47% receiving research payments.

This also contrasts with the FDA, said Prasad: "Zero percent of FDA staff has financial ties to industry."

Zero percent of FDA staff has financial ties to industry. Dr Vinay Prasad

The study authors are worried about conflicts of interest: "The presence of conflicted physicians has been shown to lead to more optimistic conclusions regarding disputed practices."

When the NCCN makes recommendations beyond an FDA approval, "it most commonly does so by removing a prior treatment, removing inclusion criteria, or expanding treatment to a different malignancy," report Prasad and colleagues.

The "biggest thing" the NCCN is doing is dropping inclusion criteria, said Prasad in an interview. He summarized the meaning of the change: "The drug is approved, but you don't have to use it just like they do in the [key] study."

Notably, at the request of the BMJ, the investigators extended their study 21 months from its "lock" point (in March 2016, when all FDA and NCCN documents were downloaded for analysis) and found that 6 of the 44 additional NCCN indications eventually received FDA approval (4 of the 6 were based on randomized clinical trial data).

Nonetheless, the authors conclude pessimistically, "Given that NCCN endorsement is linked to reimbursement by many commercial insurers and the CMS, our results suggest that payers may be covering cancer drugs with varying and scientifically less robust justification."

More than money is at stake, say Prasad and colleagues. "Low quality of evidence supporting oncology practices has led to several noted medical reversals," they write, referring to widespread interventions that are later found to be no better than previous care. 

In a statement issued today, Robert Carlson, MD, NCCN's chief executive officer, said that "CMS and other major  providers have designated NCCN as the leading source for arbitrating oncology drug and biologic coverage because of the NCCN Guidelines' proven track record for helping physicians to prolong their patients' lives and reduce their suffering."

In 1993, Anticancer Drugs Were Mostly Chemo

NCCN guidelines are used globally but are especially influential in the United States. They are one of five official compendiums for private insurers and CMS coverage.

In 1993, as part of the Omnibus Budget Reconciliation Act, Congress mandated that CMS use expert compendiums to determine coverage decisions and payment for off-label drugs used in cancer care.

At the time, this was sound governing, suggest the authors.

"The use of compendiums to justify coverage was once rational. In the early 1990s, nearly all anticancer drugs were relatively low cost, cytotoxic treatments, which were used in diverse malignancies based on some evidence," they write. Thus, no company could be expected to run and pay for new trials for new indications for such inexpensive chemotherapies.

But times, drugs, and profits change.

Now, newly approved cancer drugs typically cost more than $100,000 per year of treatment and are highly profitable; additional trials should be run when an extra indication is in question, suggest the authors. But the NCCN compendium may be decreasing incentive to do so, they stress.

One of the study's limitations is that it did not assess which recommendations were NCCN level 1, 2A, 2B, or 3, which may have implications for coverage decisions.

The authors explain that different insurers have different thresholds for coverage. For example, United Healthcare and Aetna cover all recommendations higher than 2B; Cigna covers all recommendations 2A or better and decides 2B recommendations on a case-by-case basis.

However, this is not significant, suggest the study authors, because other research has shown that 89% of NCCN recommendations are level 1 or 2A, 99% are level 2B or better, and just 1% are level 3, which is typically not covered and reimbursed.

Prasad has disclosed no relevant financial relationships . 

BMJ. Published online March 7, 2018. Abstract