Trial Tuesday: The data almost cooperated. Almost.

Here is a thought experiment: where would we be if the FDA and other regulators did not allow trial flexibility?

Keytruda, still the world's best-selling drug, got its initial US approval on a surrogate endpoint. Without that flexibility, it reaches the market at least a year later, and thousands of patients go untreated in the interim. The accelerated approval pathway is not an abstraction. It is the reason some drugs exist commercially at all.

There are enough success stories and enough blunders to keep this genuinely contested. Surrogate endpoints and accelerated approvals have become more controversial than they were a decade ago. That controversy is now influencing how companies design studies and sequence their launch plans. This week's data shows what that looks like in practice.

Every day we have a different lens. Today's is Trial Tuesday, where over the next 10 minutes we will unpack what the numbers actually said, what the industry did with them, and what both signal about where the science is heading.

There is also a feedback link at the end. Use it.

Regenxbio put forward what analysts called a strong Phase III case for accelerated approval of its Duchenne muscular dystrophy gene therapy. The data were clearly positive. They were also built on a surrogate endpoint, in an open-label trial with no comparator arm. The Rockville, MD-based company has cited prior FDA feedback as grounds for confidence that this will be sufficient. The question is whether the agency that reviews the BLA next year is the same one that shaped that guidance. In the current regulatory environment, that is not a safe assumption.

The surrogate endpoint thread runs directly into Biogen's decision to advance diranersen, its tau-targeting antisense oligonucleotide, into Phase III. As my colleague Mandy Jackson wrote for Scrip, the CELIA Phase II missed its primary endpoint, an assessment of dose response based on change from baseline on the Clinical Dementia Rating-Sum of Boxes. What it produced instead was reduced tau biomarker levels and a positive prespecified exploratory cognitive signal. Biogen pressed forward.

The data were a reminder of how hard Alzheimer's drug development remains, but also a sign that tau-targeting therapies may be advancing in ways earlier failures did not allow.

THE CALL: Study design is now a competitive signal as much as a regulatory one. How a company structures its pivotal trial tells the market something about how much it actually believes in the underlying biology. Both Regenexbio and Biogen have placed a bet. The FDA's response will be the more instructive data point.

Monday's edition flagged the scale of large-cap dealmaking in China. The clinical dimension of that story is equally important.

Our reporters trawled social media for hours and stumbled upon a conversation on Reddit about datasets generated in China being somehow inferior to those generated in, say, the US or Europe. Maybe that was true a few years ago, but not anymore.

Merck's decision to license assets from four China-based companies in a single week reflects something different. China's speed and capability in early clinical development are now globally recognized, and the data are traveling.

Christiana Bardon, a managing partner at investment firm MPM BioImpact, in a conversation with me last year, put it plainly: many of the most promising assets coming out of China have seen their data replicated perfectly in Western studies. "The discount that we used to give to Chinese clinical data," she said, "is less and less and less because the quality has become so good."

That shift matters beyond the dealmaking optics. It means the scientific rationale for China's partnerships is now as strong as the commercial one.

In related early-stage news, Taiwan-based GlycoNex is advancing an anti-glycan ADC into Phase I studies in gastrointestinal cancers, Dexter Jie Yan reported in Scrip. The agent targets glycan structures broadly expressed across multiple tumor-associated membrane proteins, an approach designed to address tumor heterogeneity in a field that has frustrated developers for two decades. CEO Mei-Chun Yang is hoping for meaningful response rates in gastric cancer from the dose-escalation stage.

THE CALL: The quality discount on Chinese clinical data is gone. The BD case for China partnerships was always about speed and cost. It is now also about scientific credibility. Those are different arguments, and the second one is harder to dismiss.

In a bid to catch up with Bristol Myers Squibb's LAG-3 inhibitor Opdualag (nivolumab plus relatlimab), Regeneron had paired its own LAG-3 asset fianlimab with PD-1 blocker Libtayo (cemiplimab). On May 15, the company reported that the 1,546-patient Phase III trial missed its primary endpoint in first-line unresectable or metastatic melanoma. A high dose of the combination produced a median PFS of 11.5 months against 6.3 months for Keytruda, a numerically meaningful gap that did not cross the statistical significance threshold (hazard ratio 0.845, p=0.0627).

The LAG-3 class has earned its skeptics. Merck halted all remaining Phase III studies of its LAG-3 asset favezelimab in 2024 after a string of late-stage failures. Opdualag, the only approved product in the class, sustained its own Phase III failure earlier this year in the adjuvant melanoma setting. The hypothesis that LAG-3 inhibition adds meaningful benefit on top of PD-1 blockade is not closed. It is just increasingly expensive to keep open.

For Regeneron, the read-across was always going to travel further than the asset. This was the first of several anticipated readouts as the company pushes to build a pipeline beyond Dupixent and Eylea, and investors were watching accordingly. BMO Capital Markets analyst Evan Seigerman was direct: "We find it hard to reconcile misstep after misstep. A big miss leaves a tough road ahead, yet again." That is not a scientific judgment. It is a capital markets one, and right now it carries more weight.

THE CALL: The LAG-3 thesis has narrowed but not closed. The class needs cleaner patient selection or a better combinatorial rationale. For Regeneron, the more pressing question is whether the next readout changes the pipeline narrative, or extends it.

AstraZeneca chalked up another Phase III win for Imfinzi (durvalumab) this week, this time in muscle-invasive bladder cancer, in combination with Pfizer and Astellas's nectin-4 ADC Padcev (enfortumab vedotin). As Kevin Grogan reported for Scrip, perioperative treatment with the combination significantly extended both event-free survival and overall survival versus standard of care. One in four bladder cancer patients has MIBC, and up to half are ineligible for cisplatin-based chemotherapy. That is a large population with limited alternatives.

The result lands against a specific competitive backdrop. The Padcev-Keytruda combination already holds accelerated approval in perioperative MIBC for cisplatin-ineligible patients, and an FDA decision on expanded approval covering cisplatin-eligible patients is expected soon. AstraZeneca's win is real. How much of the addressable population it actually contests depends on what happens next door.

What makes the Imfinzi bladder story worth the wider attention is the arc. The drug's first accelerated US approval in urothelial carcinoma came in 2017. The DANUBE Phase III failure in 2020 looked like a ceiling; AstraZeneca withdrew the indication the following year. What followed was a systematic rebuild, combination by combination, indication by indication, that has produced first-quarter 2026 sales of $1.70bn, up 34% year-on-year, with bladder and gastric cancer now driving growth alongside its established lung cancer franchise.

THE CALL: The Imfinzi trajectory is one of the cleaner case studies in how to reconstruct a drug's commercial identity after a high-profile failure. The label AstraZeneca is building now is more defensible than the one it lost in 2020, precisely because it is built on harder evidence in better-defined populations.

Medtech's IVL Gold Rush: Intravascular lithotripsy has gone from niche technology to a full-scale medtech land grab in under two years. As Dr. Phalguni Deswal reported for Medtech Insight, the dealmaking logic is straightforward: the technology uses acoustic pressure waves to fracture calcified arterial plaques, a problem every major cardiovascular player needs to solve.

Johnson and Johnson's $13.1bn acquisition of Shockwave Medical in 2024 set the pace. Boston Scientific paid $664m for Bolt Medical last year. Stryker announced plans to acquire Amplitude Vascular Systems in April, formally entering the space. Multiple trials are now running across the field.

The question is no longer whether IVL becomes the standard of care in calcified coronary and peripheral artery disease. It is who owns that standard when it does.

Three things the data said this week: surrogate endpoints are a bet, not a shortcut. LAG-3 needs a better hypothesis before it needs another trial. And a Phase III failure is not a ceiling if you are willing to do the work. The FDA will have the final word on the first. Regeneron on the second. AstraZeneca already answered the third.

Would love to hear your thoughts.

We'll be back tomorrow with Watchdog Wednesday.