This is the online edition of ESMO in 30 Seconds, a pop-up newsletter from the European Society for Medical Oncology's annual meeting in Barcelona, Spain. Sign up for the remaining editions here.

The applause mentioned above was for AstraZeneca's Phase 3 NIAGARA study in bladder cancer, which showed that giving patients the checkpoint inhibitor Imfinzi -- both before and after surgery -- cut the risk of death by 25%.

"I think they liked that one," Rebecca Dent, an oncologist at the National Cancer Centre Singapore, who was chairing the session, said as the applause finally ebbed.

With the positive outcome, Susan Galbraith, AZ's head of oncology R&D, told STAT the company plans to bring the data to regulators.

But, to get wonky for a second, there's a regulatory issue that could come into play with the drug. And that has to do with the trial design. Essentially, the FDA is growing impatient with clinical trials that test a drug both before and after surgery but don't separate out the contributions of each phase to the overall results. It's not clear whether that will be an issue for the NIAGARA study, and AstraZeneca seems confident that its data profile will carry the day. But it's an issue that scientists discussing the trials raised throughout ESMO.

More details here.

Merck partnered with Eisai back in 2018 to expand the latter company's Lenvima into new cancer indications, often testing it along with Keytruda.

It hasn't always worked. The combo approach has failed in a number of settings in lung, colorectal, and head and neck cancer.

But on Saturday at ESMO, researchers showed that adding the combination to standard of care in a type of liver cancer reduced the risk of progression or death by a third. Patients in the drug arm saw a median progression-free survival of 14.6 months, versus 10 months for those who received standard of care plus placebo -- what Josep Llovet of the University of Barcelona, who presented the data at a presidential session, called a "profound difference."

Researchers have not yet been able to calculate overall survival for the LEAP-012 study, which focused on hepatocellular carcinoma, or HCC. Eliav Barr, Merck's chief medical officer, told STAT that the companies are likely going to wait for those data before talking with regulators.

HCC is typically the result of a prior hepatitis B or C infection, or of alcohol use. It's generally treated with transarterial chemoembolization, or TACE, which involves delivering chemotherapy to the portion of the liver with the tumor while also cutting off the tumor's blood supply. But researchers have been exploring immunotherapy or immunotherapy combinations to expand the benefits of TACE.

In the trial, patients in the treatment arm saw far higher rates of side effects, with 71.3% having an adverse event that was rated severe or worse, versus 31.5% for those who got TACE alone. Researchers and Barr framed that difference as in line with what's already known about the side effects of Keytruda and Lenvima.

Lenvima works by preventing blood vessels that the tumor uses to spur its growth from forming, effectively starving out the tumor. And its success when combined with Keytruda might have to do with whether the tumors in question depend on new blood vessels for their survival. The combination has also been effective in kidney and endometrial cancers.

"All these tumors have a commonality that they have a tremendous vascularization -- the tumor stimulates blood vessel growth, that's how it keeps itself growing," Barr said about the cancers in which the drug combination has worked.

Angela Lamarca, who was the discussant following the presentation, raised the question of whether she thinks adding the drugs to TACE should be the new standard of care even without the overall survival benefit being demonstrated yet.

"My personal opinion is that, yes, I would probably recommend" it to patients, said Lamarca, an oncologist at Fundacion Jimenez Diaz University Hospital in Madrid.

Coming into ESMO, Bristol had already announced that it would be moving the combination of its PD-1 drug Opdivo and the LAG-3 targeting relatlimab into a Phase 3 trial against Merck's Keytruda in a form of non-small cell lung cancer, with both arms also using chemotherapy. The patients enrolled in the RELATIVITY-1093 study would have non-squamous cancers and intermediate levels of PD-L1 expression on their tumors, from 1% to 49%.

The company's justification for that criteria was presented here on Saturday. In the Phase 2 RELATIVITY-104 study, researchers looked at the responses to the treatment by groups of patients -- based on squamous vs. non-squamous cancers, as well as on levels of PD-L1 expression. Bristol said that it had designed the study to figure out which subgroup within NSCLC benefited from adding relatlimab to a backbone of chemotherapy and a PD-1. (Bristol's Opdualag, which is approved in melanoma, is also a combination of relatlimab and Opdivo, though the combination was used at a higher dose in the lung cancer study.)

"That was exactly the answer we were looking for and what we got," Samit Hirawat, Bristol's chief medical officer, told STAT ahead of the conference.

Indeed, the selected subgroup that received the combo of immunotherapies saw a response rate of about 60% and a progression-free survival of 9.8 months at the median, far outperforming those who received only Opdivo and chemotherapy, who saw a response rate of less than 30% and PFS of less than six months.

But the Phase 2 trial design, as well as those results, came under some notable criticism at the conference. The discussant at the session, Marina Chiara Garassino of the University of Chicago, said that researchers were "cherry-picking" by weeding through the subgroups in the way they did. She also noted that past studies of PD-1 drugs had shown progression-free survival of about 9 months in patients with intermediate PD-L1 expression levels -- not far off from the 9.8 months the Bristol combination reached.

"We don't see that there is a big improvement in terms of progression-free survival" with the addition of relatlimab, she said.

Writing in ApexOnco, Jacob Plieth argued that problems with the Phase 2 results suggest "why it's taken this long for Bristol to reveal its data (results had once been expected in 2022), and accordingly why RELATIVITY-1093, a pivotal study first mooted all the way back at the start of 2021, is only now getting under way. Presumably Bristol has only now found sufficient backing to start RELATIVITY-1093, but the ESMO data cast doubt on such a view."

Should be an interesting Phase 3.

Read more here.

Experts love to tell reporters that they shouldn't compare results from different trials. There can be different patient populations, and different statistical methods, and different timelines. It's all a bit fraught.

But you know what experts love to do? Compare trials.

This "is something that we are always told not to do," Lamarca, the discussant on the liver cancer trial mentioned above, said as she showed a cross-trial comparison.

We're not picking on Lamarca. We've seen a number of discussants do this, often while giving some line about how you're not supposed to do it. The hullabaloo caused when a discussant compared trials at the European Society of Cardiology conference a few weeks ago is also fresh in our minds.

Of course, wanting to compare trial results make a lot of sense. When clinicians are trying to weigh the respective risks and benefits of different interventions, it's one of the simplest strategies they have, particularly given how few head-to-head studies there are.

So don't get mad next time a reporter does it. As Lamarca said, "I believe it's important to put this into context."