On Tuesday, Merck reported better-than-expected first-quarter profits, driven by a huge jump in sales of its cancer drug, Keytruda. 

Merck has high expectations for the next-generation oncology treatment, and justifiably so. Last month, it presented slam-dunk data at the American Association of Cancer Research (AACR): in a Phase III combination trial, Keytruda plus chemotherapy decreased the risk of death by 51% for newly-diagnosed patients with the advanced, non-squamous type of non-small cell lung cancer (NSCLC) compared to chemotherapy alone. 
The combination treatment also showed statistically significant benefits for stalling the progression of the disease.
Taken altogether, the results were “groundbreaking,” said David Querry, president of Navicor, an agency focused on the oncology market. “What it suggests is that chemotherapy alone is no longer a standard of care,” Dr. Leena Gandhi, a leader author of the study and director of the Thoracic Medical Oncology Program at the Perlmutter Cancer Center at New York University Langone Health, told The New York Times
Lives hang in the balance: Lung cancer is expected to kill more than 154,000 Americans this year. As do revenues: In a recent note to clients, Goldman Sachs said Keytruda has the potential to generate $16 billion in sales by 2025. Merck’s stock has steadily grown in value since early April. 
Meanwhile, rival Bristol-Myers Squibb’s share price has trended in the opposite direction. The New York City-based pharmaceutical company also presented data from a trial at the AACR, which combined its immune-boosting drug Opdivo with Yervoy, an older immune-boosting drug. While the trial met its co-endpoints, the findings were less impressive, and only worked in a segment of the study’s patients: those whose tumors had high expression of a biomarker called tumor mutational burden (TMB). 
BMS posted better-than-expected first-quarter earnings last week, driven in large part by sales of Opdivo. But investors remain skeptical that it can defend against Merck’s looming dominance in the lucrative NSCLC market. 
To quote Credit-Suisse’s succinct analysis: “MRK the Clear Winner at AACR, While BMY Will Need Time to Catch Up.”
As its R&D plan plays catch up, the ability of BMS to make a commercial impact depends on the extent to which the company can convince physicians of the advantages of the I/O-I/O combination vs. I/O-chemo. Here’s a more detailed look at how the dueling I/O drugs got to this point, and how events could progress from here. 
Merck, BMS will continue to swap roles
Opdivo and Keytruda were both approved to treat melanoma in 2014. The PD-1 checkpoint inhibitors represented a novel strategy in the war against cancer: harnessing the body’s immune system to attack malignant cells. 
While the therapies are similar, Bristol-Myers’ and Merck’s strategies for entering the NSCLC market differed. Merck took a more conservative approach, focusing on patients with high PD-L1 expression, or those most likely to benefit from disabling PD-1 immune system cells. Meanwhile, BMS targeted patients regardless of their PD-L1 status. 
At first, this worked well for BMS. Opdivo was approved as a second-line treatment for all lung cancer patients, while Keytruda was approved only for those with high levels of PD-L1 expression. But as Keytruda racked up a series of clinical trial wins, this dynamic began to shift. If there was an inflection point, it may have been when Opdivo notably failed to meet its clinical endpoints as a monotherapy in first-line treatment. 
The latest results out of both companies should accelerate this transposition. Keytruda in combination with chemotherapy showed a clear benefit in terms of lengthening the survival time of patients with non-squamous NSCLC who exhibit high PD-L1 expression — as well as those with low PD-L1 expression. The BMS drug combination, meanwhile, showed a benefit in terms of keeping patients’ cancer from worsening among those who exhibited high TMB, which is an independent biomarker from PD-L1 status.
In other words, the dynamic “flip-flopped,” Querry explained. The new data coming out of Merck indicates the combination could work on a far broader percentage of patients, while “BMS is shifting towards a different biomarker in a clearly defined patient population.”
Merck expands access to NSCLC patients — effective immediately 
In 2017, Keytruda in combination with chemotherapy received accelerated approval as a first-line treatment for NSCLC patients regardless of PD-L1 status, but the decision was based on one cohort of a phase II trial.
The new, more robust data from the Keytruda combination study strongly supports that the “regimen is effective and can be immediately applied to practice,” said Megan Epperson, an oncology analyst at the consulting firm Kantar Health.
The combination therapy could also potentially target other biomarkers, including TMB. Merck has the “opportunity to expand the marketplace by separating [Keytruda] from PD-L1 expressions,” Querry said. “From a marketing standpoint, that’s what I would be looking at.”
BMS targets specific segments of lung cancer patient — but first, it needs more complete data 
Bristol-Myers’ combination trial was successful in that patients with a high mutational tumor burden who received a Opdivo-Yervoy combination regimen were 42% less likely to see their disease worsen than those receiving standard treatment. But it’s too early in the trial to show an overall survival benefit. 
Unfortunately for the company, these caveat-laced results don’t exist in a vacuum. Given Merck’s “groundbreaking” data for Keytruda, BMS has been doing a lot of explaining, and will need to continue to do so if it is to keep pace with Merck. 
In light of a dearth of data on overall survival of this combo, wrote Credit-Suisse analysts, “It may come down to BMY being able to effectively market the message that the I-O/I-O combo would allow patients to avoid chemotherapy and/or save chemo for potential use in a second line setting.” 
Indeed, this has been a major talking point for the company. For physicians, “it’s definitely an important factor,” Epperson said. Without more mature data, however, the talking point is largely theoretical. 
The two trials should not be compared head-to-head, as they contain “a number of differences in terms of pathology, trial design, and biomarkers,” Epperson said. 
But in real-world settings, clinicians will make such comparisons, Querry argued, which means Keytruda could benefit from a “cross-study comparison and halo effect.”
“Anything Opdivo does, I think there is going to be a natural assumption that Keytruda could do better,” he predicted.