With New Data, Drug Combos, Lung Cancer Experts Scramble to Keep Pace

Xconomy National — 

When oncologist Renato Martins finished his medical training, advanced lung cancer was almost certainly a quick death sentence. “I knew, by name, every patient who had survived two years,” he says.

Thanks in large part to the arrival of cancer immunotherapy, things are much different today. While lung cancer remains by far the leading cause of cancer death, immunotherapy now offers patients with advanced disease—Stage 4 cancer that has spread to other parts of the body—at least the chance to live longer than previously thought possible.

Here’s a recent example. In one study, 16 percent of 129 patients receiving an immunotherapy were alive after five years. That low percentage might not sound encouraging, but the typical five-year life expectancy for patients with non-small cell lung cancer, or NSCLC—the most common form of the disease—between 1999 and 2010 was about 10 percent for some Stage 4 patients—and less than 1 percent for others whose cancer had spread widely, according to the American Cancer Society.

For Martins, the medical director of thoracic/head and neck oncology at the Seattle Cancer Care Alliance, it’s a big deal.

“We’re talking about five-year survival for patients with advanced lung cancer,” he says. “Things have changed dramatically.”

The cost of cancer care is also changing, but even with national concerns over healthcare costs, Martins and others have yet to see pushback from payers. Immunotherapies on their own cost around $13,000 per patient, per month, or roughly $150,000 per year. That means big money for their developers: Last year, Bristol-Myers Squibb’s (NYSE: BMY) nivolumab (Opdivo) generated nearly $5 billion in sales, and Merck’s (NYSE: MRK) pembrolizumab (Keytruda) pulled in almost $4 billion.

More change looms. Study results released earlier this week could make pembrolizumab a go-to drug even without chemotherapy in previously untreated patients. And in a few days, three more studies in similar patient populations will be featured at the American Association for Cancer Research’s annual meeting in Chicago. Each could add more options for people newly diagnosed with advanced NSCLC. These studies are known as Keynote-189 (from Merck), Checkmate-227 (from Bristol-Myers Squibb), and IMpower150 (from Roche); all are testing immunotherapies known as checkpoint inhibitors combined with treatments such as chemotherapy.

While doctors and patients wait for those combination results to become public until Monday, they have other news to digest. Merck announced two days ago that advanced NSCLC patients receiving just pembrolizumab lived “significantly” longer than those on chemotherapy in its Keynote-042 study. (An important side note: The drug was effective on patients with various different levels of the protein PD-L1 expressed by their tumors. Pembrolizumab and other checkpoint inhibitors are designed to block PD-1, which helps tumors hide from the immune system. But PD-L1 tumor levels have been an inconsistent and frustrating “biomarker”—that is, a way to predict who might respond to immunotherapy.)

Merck still needs to fill in details of Keynote-042, but it could lead to a new way of thinking about treatment, says Martins, putting immunotherapy first and making chemotherapy a potential supplement for newly diagnosed patients.

That’s important because many patients are too fragile to withstand chemotherapy, says Young Kwang Chae, co-director of the clinical trials unit at the Lurie Cancer Center in Chicago and an assistant professor of medicine at Northwestern Medicine. The side effects of checkpoint inhibitors can range from drug to drug but have mainly been tolerable for patients. (The high prices, however, have contributed to what is widely seen as unsustainable spending on U.S. healthcare—particularly as they are combined with other treatments.)

Lung cancer specialists interviewed by Xconomy are eager to see the details from these studies and expect when the dust clears to have a wider range of treatment options to sort through.

“It’s really breathtaking,” says John Heymach, the chair of thoracic/head and neck medical oncology at the University of Texas MD Anderson Cancer Center. “It’s going to take a while for the field to figure out the optimum regimen to use in different patient populations.”

For decades, chemotherapy was the main option for patients with advanced lung cancer. Then came targeted medicines, which hone in on mutations in tumor genes like EGFR and ALK and are often easier to tolerate than chemotherapy. But most lung cancer patients don’t have these mutations, and the large remaining pool is where immunotherapy has begun to make its mark.

First, in October 2015, the FDA approved nivolumab and pembrolizumab within days of one another for patients whose NSCLC had spread after chemotherapy. Then in 2016, pembrolizumab succeeded in a Phase 3 trial in newly diagnosed advanced NSCLC patients, but nivolumab failed. Pembrolizumab became the first FDA-approved immunotherapy to treat lung cancer with abnormally high PD-L1 levels. Merck followed with an FDA approval of pembrolizumab plus chemotherapy in newly diagnosed patients. (Keynote-189 is a larger, longer study meant to confirm the earlier pembro-chemo results.)

Despite all this progress, immunotherapies still only work in roughly 20 percent of patients. The race is on to figure out why and do better, which brings us to the upcoming AACR presentations. Data from Checkmate-227, Keynote-189, and IMpower150 are all due Monday morning.

Xconomy asked lung cancer specialists who aren’t involved with those trials what they’ll be looking for.


As noted, PD-L1 is one immunotherapy biomarker. Another is “tumor mutation burden,” or TMB, a measure of the level of genetic mutations in the tumor cells. Checkmate-227 is all about TMB. Bristol said earlier this year that a combination of its two checkpoint inhibitors, nivolumab and ipilimumab, kept cancer from spreading better than chemotherapy—at least in patients whose tumors have high TMB. The picture should fill in with much more detail in coming days.

Doctors want to know about the toxicity of the combination; ipilimumab on its own has a track record of more severe side effects. They also want to compare the combination to nivolumab alone and—even though comparisons across trials are always tricky—to pembrolizumab alone or pembrolizumab plus chemotherapy. Survival rates are important. Chae of Northwestern wants to know if the nivolumab/ipilimumab combination is “just buying a few more months and then everyone progresses.”

“Response rate and progression-free survival are important, but overall survival is always the most important endpoint,” he says.

Heymach of MD Anderson says good results could make the ipilimumab-nivolmuab combination a preferred option for one type of lung cancer: squamous cell carcinoma with low PD-L1 levels and high TMB.

Even if these questions turn up positive for Bristol, Chae has another concern. DNA sequencing is required to calculate TMB and can take a few weeks. Will oncologists and their patients newly diagnosed with advanced lung cancer be willing to wait until they can start treatment? “I think with time and advancements in technology this will be solved eventually,” Chae says, but “at the moment I can easily see that becoming an issue.”


When the FDA approved the pembro-chemo regimen in May 2017, it did so based on a small randomized study, Keynote-21G, that didn’t have the drug’s effect on overall survival as a main goal. The 616-patient Keynote-189 will have survival results: How much longer people receiving the combination live than people taking only chemo. The magnitude of that effect will be telling, says Heymach of MD Anderson. If it’s significant, it could “cement chemotherapy plus pembrolizumab” for NSCLC patients who have lower PD-L1 levels and don’t have squamous cell carcinoma.

Martins is watching for another key detail from Keynote-189. Patients receiving chemo alone were allowed to add pembrolizumab to their treatment if they progressed. If the crossover rate is high, that bolsters the idea that providing the combination early produces better results than waiting. “I think that will be a pretty big deal,” he says.

The wild card with all this, however, is the Keynote-042 study that Merck partially released earlier this week. Again, the hint is that pembrolizumab alone could be better for first-line patients than pembrolizumab combined with chemo. If those tantalizing results hold up in detail, Keynote-042 “changes everything,” Martins says.


Roche’s combination is the most aggressive and complicated of the new studies. In addition to the immunotherapy atezolizumab, it includes the Roche drug bevacizumab (Avastin) that has helped scores of patients with a variety of cancers since it was first approved in 2004. The mix also includes chemotherapies.

So many drugs at once gives the oncologists we interviewed some pause. Toxicity is one concern; Chae notes that Roche is using the chemotherapy paclitaxel instead of pemetrexed even though paclitaxel is associated with more frequent side effects like peripheral neuropathy, which isn’t associated with pemetrexed. There’s also “financial” toxicity: “That is a pretty pricey therapy,” Martins says. “At some point we will not be able to afford it.”

But the early details on IMpower150, released in December, gave oncologists a hint that chemo, immunotherapy, and bevacizumab could work well together. It also held out hope for patients with a mutation in the EGFR gene who haven’t fared well on targeted drugs like erlotinib (Tarceva).

“That’s probably how I would take advantage of IMpower150,” Chae says. “I think it has its role.”

Looking Ahead

No matter how the data play out in coming days, the pace of change is “overwhelming,” says Chae.

With tumor mutation burden now coming to light as an important diagnostic measurement, for instance, lung cancer is becoming ever more stratified. Chae is looking forward to each study’s analysis of patient subgroups, not just the overall numbers. “It will truly be a personalized approach with this many options coming aboard,” he says.

Who will be able to afford personalized cancer medicine? Heymach, Martin, and Chae all say they’ve had no problems getting patients access to immunotherapies so far.

But as many observers have noted when competition doesn’t seem to bring down the cost of similar drugs, Heymach says, the market “isn’t operating as other markets commonly do.”

With combinations that pile on the drug costs, will insurers finally push back? Add to that the cost of the diagnostic tests—genomic sequencing and the like—to predict patients’ responses. For a lung cancer patient, are payers “going to pay for mutation testing and TMB testing and PD-L1 testing?” Heymach says. “Will they specify one of these regimens, or will they leave it open for any of these regimens for patients meeting the criteria? That’s something that we don’t know yet.”