Payer Testing Delays Can Lead to Patients Taking Wrong Therapy, Panelists Say

The goal of precision medicine—to get “the right drug to the right patient at the right time”—is derailed when patients and their doctors hit a roadblock: insurers will not approve molecular genomic tests, or they will not do so without long prior authorization (PA) delays, which leave a window for cancer to advance.

This persists, despite trials that have revealed dozens of new targets across multiple cancers, especially in lung cancer, which still accounts for most cancer deaths. Evidence now shows that testing is not simply recommended, but that failure to do so can cause patients to miss out on a targeted therapy—or worse, they may start chemotherapy that will limit the effectiveness of innovative treatments down the road.

Oubre

This message came through Monday during the Community Oncology Alliance (COA) Payer Exchange Summit, held in Reston, Virginia, where a diverse panel of stakeholders grappled with what PA delays and coverage limits look like in the real world. The discussion, “Understanding Biomarkers and the Role They Play in Oncology,” featured moderator Kathy Oubre, MS, CEO of Pontchartrain Cancer Center in Louisiana, and the following panelists:

• John Fox, MD, senior director for the Americas, Illumina

• Bret Jackson, president and CEO, Economic Alliance for Michigan

• Scott Kruger, MD, Virginia Oncology Associates

Oubre first shared her recent experience with the largest payer in her area, Blue Cross and Blue Shield of Louisiana (BCBSLA).

“We’re thrilled to have these cutting-edge [immuno-oncology] therapies, and they’re really transforming the way we are able to treat our patients,” she said. “But then it got complicated very quickly.”

When it comes to testing, “Most of our payers want to choose for us,” she explained, and even within a single company, forms for individual tests can vary. The process is not integrated into the electronic medical record, so ordering next-generation sequencing (NGS), waiting for PA, and then waiting for the results slows down patient care.

In February, BCBSLA “determined that all NGS would go into its own prior authorization period of 10 to 14 business days. They deemed it experimental,” Oubre said. Medical records must be sent before the clock starts. Usually approval happens, but the chain of information is not always smooth. “When we’re talking about barriers to access, that’s a huge problem,” she said.

After taking on BCBSLA, Oubre said the insurer changed its policy, which affects about 25% of the patients her practice sees.

Kruger has also seen improvement, but it’s also been a slow climb. Five years ago, some insurers required oncologists to order biomarker tests one at a time—it took 6 weeks to get a full lung panel.

Today, he said, the testing companies have taken up the fight. “We have stopped getting approval beforehand, because it takes too long—it can take up to 4 weeks in my community. We just send it and allow the companies to get the approval, and they’re pretty good at it,” Kruger said. Under this model, the out-of-pocket exposure for a patient is limited. “It usually works out, but it doesn't work out as smoothly as it should.”

Fox

Fox said the landscape remains complicated. “Most payers will at least cover a multigene panel of less than 50 genes,” he said. That sounds like a lot, but there are 23 different panels on the commercial market, with some focused on lung cancer and others generally on solid tumors. Do payers know which of the commercially available panels includes the 12 ranking biomarkers under the National Comprehensive Cancer Network guidelines? It turns out they don’t.

“When I show them this, they’re stuck,” Fox said. “They’ve never done the analysis.”

So, what do oncologists do? They can order a large panel that might cover more genes, but it might take weeks to get approval—and then the clock starts ticking on the 14 days to get back test results. Or they can select a smaller panel with greater chances of approval that might miss some important biomarkers.

In the meantime, Fox said, many oncologists will start patients on chemotherapy or perhaps pembrolizumab in non–small cell lung cancer, because it’s indicated for all PD-L1 indications.

“And what’s going to happen? Fifty percent of those patients who don’t have results will be indicated for a targeted therapy,” he said. Emerging evidence shows that patients who start on pembrolizumab and later switch to targeted therapy have worse overall survival than those who take a targeted therapy from the start.

“But the problem is, patients want to get treated. So, you can you get started on the wrong therapy.” For employers, he said, this should be exasperating and shocking news—for want of a timely test, patients could have avoided an expensive therapy such as pembrolizumab in favor of a targeted therapy that would have been the better choice—and the likely better option in the long run.

Oubre noted, “It’s not just the employer paying into that. The patient also has additional co-pays—and taking chemotherapy is not an optimal treatment while they are waiting.”

Jackson said Michigan’s situation is different, as the 3 top insurers that cover 85% of the population are based in the state, and this increases accountability. There are legal requirements for assays to be done in 14 days or less and most testing PAs are done in 1 to 2 days.

Fox said despite some successes, there remains “functional illiteracy” among many payers on these issues, which contributes to patients getting the wrong drugs. Oubre recommended use of the biomarker toolkit from the National Cancer Treatment Alliance, which she said was useful in her talks with BCBSLA. She asked the panelists to focus on exactly what the barriers are to gaining access to testing.

Fox and Kruger had an exchange about the lack of quality standards surrounding testing. Fox said the Personalized Medicine Coalition did a point-in-time analysis that found only 50% of patients with cancer were being tested and only 36% ended up getting the right therapy. “The reality is that most patients aren’t getting tested,” he said. “We think things are getting better, but we don’t really have a good measurement.”

Kruger agreed, noting that HER2 testing in breast cancer remains an exception. He said one problem is that the question of who is responsible for testing remains unsettled—it can be the hospital, it can be an outside laboratory, it can be the community oncologist. Until recently, he said, medical oncology has not done a great job with making sure testing occurs.

Fox said for all the talk about patient-centered care and the patient experience, “Patients don’t give a rip about the patient experience if you’re not on the right drug. The problem is, they have no idea.”

He continued, “I appreciate that all the systems are different. But in the absence of ongoing measurement, we don’t really know how any individual system is doing.”

Jackson offered a long anecdote that demonstrated why leaving this responsibility to the employer is not always feasible, because a benefits administrator overwhelmed with too many other responsibilities is not going to have the bandwidth to understand the intricacies of making sure patients with cancer gain access to genomic testing.

Fox, however, felt that the employers’ interest as the one paying the bill for cancer care is key.

“There's that old saying that not everything that can be measured matters, and not everything that matters can be measured. That's not true for molecular diagnostic testing,” he said. “I think we have to acknowledge what we don't know today, at a population level, or even in practice settings.”

“We do know that measurement is the first step in improving. So, my request or my action step is to make sure that we galvanize around what appropriate measurement is,” Fox said. “But we [must] galvanize around what we need to do to ensure that patients get the right treatment at the right time, that we don't waste employer dollars. And we don't cause harm by providing the wrong therapy.”

Kruger said that data show a medium-size panel is adequate for most cancer cases. “If we could change the billing, and we could all agree on getting a middle panel, that would be a very reasonable thing,” he said.

Oubre called on attendees to get involved in advocacy at the state level and pointed to COA leaders who have helped her in her efforts in Louisiana. “Several states have enacted bills around this, and it won't help everyone, but it's definitely a start.”