*|MC_PREVIEW_TEXT|*

CV Hypotheses Shrugged, Hearing HF, and Interventional Longevity
By Viktor Zarev
November 16, 2026
site logo

Together with

partner logo

“The most exciting phrase to hear in science, the one that heralds new discoveries, is not ‘Eureka!’ but ‘That’s funny…’”

Isaac Asimov

Mr. Asimov said it best, and today’s top story on the ZEUS and Lp(a)HORIZON trials follows his logic. After all, science is just as much about understanding what went wrong as it is about proving what went right.

My exceptionally talented colleague, Jack Troy, also covered these two trials last week in The Bio Wire but with his signature twist on what it means for the broader pharma industry, so be sure to check that one out too!

As always, thanks for reading,

Vik

Cardiology Pharmaceuticals

What We Learn When Big Hypotheses Fail

We’ve known for months now that neither Novo’s ZEUS trial nor Novartis’ Lp(a)HORIZON yielded positive topline results for their respective drugs and while the full data at AHA 2026 wasn’t a win for these pharma giants, it’s certain to help shape cardiac science.

  • ZEUS evaluated whether or not Novo’s monoclonal antibody, ziltivekimab, could reduce MACE in ASCVD patients by targeting IL-6 and hs-CRP to lower inflammation.
  • Lp(a)HORIZON also aimed at evaluating if pelacarsen (Novarits & Ionis) could reduce MACE by reducing Lp(a).
  • Both trials rested on respected hypotheses since inflammation and Lp(a) are two of the most-watched residual-risk targets for patients with controlled LDL-C and BP.

Even when the disappointing topline results emerged, both trials proved that pharmacologically, the drugs generously lower these biomarkers, but clinically, there was no impact on MACE.

  • Ziltivekimab lowered free IL-6 by 76% and hs-CRP by 64% with no MACE benefit.
  • Pelacarsen reduced Lp(a) by 78% in Lp(a)HORIZON and led to no MACE benefit as well.

So what happened? After all, both ZEUS and Lp(a)HORIZON were well-funded, sufficiently powered, randomized, placebo controlled, the whole nine yards — but there’s plenty of room for speculation.

  • For ZEUS’s results, some researchers offered suggestions like “nonmodifiable late-stage disease” prevented older, sicker patients from benefitting.
  • That same sentiment surfaced in comments about Lp(a)HORIZON, and whether patients’ ASCVD was too far gone to improve by lowering Lp(a).

But across the board, the most common admission was – “we don’t know.” That’s just a reality of the scientific process, and it’s becoming increasingly common in cardiac drug development.

  • For example, GLP-1 research suggests the drug class can provide cardiovascular benefits beyond weight loss, but we don’t know why yet.

What’s next after we’ve mourned ZEUS and Lp(a)HORIZON?

  • Novo has already canceled three out of the four ziltivekimab trials (ZEUS was one) and the drug now only has the ARTEMIS study for post-MI recovery to prove itself.
  • Novartis has also pulled the plug on all pelacarsen research except for the Lp(a)FRONTIERS CAVS study examining the drug’s impact on calcific aortic valve stenosis.
  • That said, other companies’ trials like Amgen’s OCEAN(a), Eli Lilly’s ACCLAIM, and Silence Therapeutics’ Phase 3 zerlasiran study are all still testing the Lp(a) hypothesis.

The Takeaway

As strange as it may seem, the failure of these two drug trials is a win for cardiac science. Even well-founded hypotheses can flounder, and while it might displease the developers and their investors, knowing where not to look can greatly accelerate the search for the right answer.

The Benefits of Outsourced Post-Processing

Using an outsourced cardiac image post-processing solution doesn’t have to mean sacrificing control of the results. Discover how PIA’s customizable post-processing workflow can help you get the most out of your images.

sponsor logo

From Fetal to Adult Cardiovascular Ultrasound

Continuity of care is critical when it comes to a patient’s cardiovascular health. That’s why cardiologists across America are picking Fujifilm’s Lisendo 880 cardiovascular ultrasound for its versatility and precise imaging starting at a patient’s fetal development to adulthood.

sponsor logo

How AI Expands the Role of CMR Experts

If you’re an MRI technologist, AI could be the best thing that ever happened to your career. Read here to learn about the benefits of AI for CMR experts and how it can accelerate both your training and opportunities.

sponsor logo

The Wire

  • Listening for HF Before It Hits: Can a patient’s voice tell clinicians that their HF is worsening? DETECT-HF says partly. Across 563 patients making daily smartphone recordings, Cordio’s HearO System caught 52 of 76 first HF events (68.4% sensitivity), an average of 24 days before they appeared clinically. While impressive, that result fell short of its prespecified sensitivity target, though its alert rate (2.65/patient-year) met its goal.
  • Lorundrostat Holds Up at 48 Weeks: Blood pressure drops with lorundrostat appear to last a while. In the Transform-HTN extension study, Mineralys’ aldosterone synthase inhibitor cut automated office systolic BP by 17.8 mmHg from baseline at Week 48 (n=504), and 56% of patients reached <130 mmHg. Hyperkalemia requiring dose changes or discontinuation occurred in 5.1% of 632 patients. It’s worth noting that this study was designed to provide durability data, not efficacy proof since it has no placebo. Lorundrostat’s FDA’s PDUFA date is December 22.
  • ARPI’s LVEF ECG Software Clearance: South Korea’s ARPI landed FDA 510(k) clearance for its ECG Buddy-LVD, HF screening software. It works by reading a standard 12-lead ECG to flag adults who may have an LVEF of 40% or less, enabling quick screening where echo isn’t readily available. The software is externally validated on more than 22k U.S. records from MIT’s MIMIC-IV database and its clearance marks ARPI’s first ever entry into the American market. 
  • Google’s Smartwatch Hypertension AI: Google Research built an AI model that flags hypertension from a consumer smartwatch’s PPG and accelerometer signals, with no cuff calibration. Validated against 24-hour ambulatory BP in 196 participants without known CVD or BP meds, the model hit 65.8% sensitivity and 90% specificity over seven days of real-world wear. That beat office BP screening’s 55.3% sensitivity. Published in Nature, it’s pitched as a nudge toward cuff-based confirmation, not a diagnosis.
  • Valve on Valve on Valve: Sometimes fixing a failing valve-in-valve means breaking the valve underneath (yeah, that means three valves in one annulus). A recent EHJ case report details how an 88-year-old woman with a degenerated SAPIEN 3 valve placed inside her SJM Epic surgical valve presented to physicians with worsening valve dysfunction symptoms. To address this, her care team fractured the surgical frame using a transcatheter balloon which also expanded the degenerating SAPIEN. They then implanted an Evolut FX+ valve inside the SAPIEN, ultimately improving her valve dysfunction symptoms.
  • Community CPR Training Doesn’t Move Survival: The state of North Carolina trained more than 200k people in hands-only CPR and cardiac arrest survival rates didn’t budge. In the RACE-CARS trial, researchers designed a CPR training program for cardiac arrest bystanders and the 911 dispatchers giving them over-the-phone instructions. Unfortunately, the counties that implemented the training saw no difference compared to those that didn’t when it came to the percentage of patients who survived CA and discharged with good neurological outcomes (9.0% vs. 9.4%).
  • ProtEmbo Hits a TAVR Wall: Casting doubt on the benefits of embolic protection during TAVR, the PROTEMBO IDE trial stopped early for futility. After randomizing 305 TAVR patients to one of three arms: Protembis’ three-vessel ProtEmbo, Boston Scientific’s two-vessel Sentinel, or no protection, researchers found that the ProtEmbo device met its safety goal (7.6% MACCE), but resulted in the highest new brain lesion volume. The trial also wasn’t powered to compare stroke rates, and its lead investigator sees no evidence to support adopting the ProtEmbo device.
  • ISCHEMIA’s Long Term Results: More than a decade out, early stents or bypass still don’t help stable CAD patients live longer. The ISCHEMIA-EXTEND study followed 5.1k patients with moderate-to-severe ischemia for up to 13 years, finding that all-cause mortality was 25.8% with an initial invasive strategy compared to 26.9% with medical therapy first. Still, procedural interventions improved quality of life for symptomatic patients, so angina relief remains the case for going invasive.
  • Merge Unifies Enterprise Image Management: Merge launched a new cloud-based enterprise image management platform called Merge Imaging that unites radiology, cardiology, and clinical imaging in a single system. Merge Imaging enables clinicians to use a single workflow for viewing and managing images, and includes applications running on the company’s MergeIQ AI orchestration layer. One of these is AssignIQ, a new AI-based tool that matches new cases to the most appropriate specialist.
  • Tape Measures Make a Comeback: A new article in JAMA rounded up the evidence supporting the push to move beyond BMI as the primary measure of obesity with metrics such as waist circumference. The authors cited a JACC study where waist size reclassified cardiovascular risk even for people with normal BMIs and a JAMA Network Open paper where using BMI plus waist to identify obesity performed at least as well as more complex definitions when compared with DEXA scan body fat measurements. 
  • AI Changes After FDA Clearance: How often do medical AI algorithms undergo changes after they get FDA clearance? A new study in JAMA Network Open sheds some light, tracking change histories for 163 AI applications (89% for analyzing medical imaging data). In all, 34% received at least one update, with a median time to first update of 1.8 years, and of these, 75% got an update that changed their usage. Many updates lacked consistent documentation on performance, indicating improvements in reporting may be needed.

A Rad’s Eye View

AI-enabled post-processing software dramatically reduces cardiac imaging analysis times while improving consistency and collaboration between radiologists and cardiologists. Hear from Laura Dragonetti, MD on how detailed CMR quantification enabled by Circle Cardiovascular Imaging detects disease much earlier.

sponsor logo

State-of-the-Art Cardiac Monitoring at Scale

Cleared for 13 different heart rhythm interpretations, AccurKardia’s AccurECG 2.0 shortens the time from an arrhythmia incident to clinical intervention. Watch here to find out how AccurKardia and Specialized Medical are teaming up to bring the AccurECG 2.0 technology to clinicians across the U.S.

sponsor logo

Integrating AI for Enhanced Echo

Sometimes it pays dividends to be an early adopter. Find out how the Christ Hospital’s early evaluation of Merge Cardio 12.5 allowed for remote echocardiography measurements, enhanced reporting accuracy, and a seamless user experience, bringing even greater efficiencies and clinical confidence.

sponsor logo

The Resource Wire

  • Meet with Optum at RSNA 2026: Whether you’re evaluating PACS replacement, cloud migration, AI readiness or enterprise imaging expansion, meet with Optum at RSNA to see the latest advancements and explore practical strategies for modernizing imaging while reducing complexity.
  • HFSA Late-Breaker Findings: New research presented as a Late Breaker at HFSA demonstrated that patients receiving CCM® Therapy experienced significant long-term improvements in heart failure hospitalization rates and other key measures of cardiac health, including NT-proBNP, functional status, quality of life, and diuretic use… Read More.
  • Canon’s 30 Years of Caring: For over 30 years, Canon has helped advance pediatric interventional cardiology through innovation and a commitment to better care. Learn more about how and why modern pediatric cardiac angiography came to be from one of its founding members, Dr. John P. Cheatham.
  • VASCADE MVP® XL Vascular Closure System: Sometimes EP procedures call for a venous vascular closure system that is simply bigger. Haemonetics’ VASCADE MVP XL closure system comes with 58% more collagen and a 9% larger disc compared to the VASCADE MVP system as well as an expanded indication for utilizing 10-14F inner diameter sheaths.
  • Discover More: About CARDAMYST® (etripamil) nasal spray
  • Pulmonary Hypertension Detection With a Single Echo Clip: Pulmonary hypertension is often diagnosed late, when irreversible vascular remodelling has already occurred. That’s changing, however, with Us2.ai’s spatiotemporal transformer technology that can process apical four-chamber echo and accurately detects pulmonary hypertension far beyond the single-view echocardiographic benchmark.
  • Most Health Systems Have Deployed AI. Few Have Scaled It. 75% of health systems have deployed GenAI, yet only 4% report scaled implementation with measurable outcomes (Becker’s 2026). Cardiology shows the difference: HCM diagnosis in about three months versus 2 to 5 years, and pre discharge ILR placements up from 3 to 51. Join Viz.ai CMO Tim Showalter, MD on October 22 to hear what it takes. Register now.
  • Beyond Heart Rate – What ECG Intervals Can Reveal: The ECG waveform contains more information than heart rate alone. For device makers and cardiac software developers, interval measurement is an important part of translating raw signals into usable data. Explore Monebo’s Kinetic Intervals technology and its role in ECG analysis.

The Industry Wire

  1. Democrats renew effort to open Medicare to people 50 and older.
  2. Mayo Clinic appoints AI leader as next president and CEO.
  3. Former FDA head Makary accuses Trump aides of bowing to industry.
  4. DC shooting suspect identified as physician, health tech executive.
  5. Who can be licensed as a physician? Not AI.
  6. FDA orders vision loss safety warnings for GLP-1s.
  7. The fight over surprise medical bills is coming for employers.
  8. U.S. loses measles elimination status. 
  9. Medicaid work requirements delayed for six months.
  10. AI health advice can be accurate and still make us sicker.