Interventional cardiologist, CV outcomes & policy researcher, trialist, sports fan, @PennMedicine, @VAPhiladelphia, @PennLDI, tweets reflect only my opinions
Philadelphia, PA
Joined December 2014
- Tweets5.7K
- Following1.4K
- Followers4.4K
- Likes6.9K
Pinned Tweet
Thread/
New @AHAScience Scientific Statement published online in @CircAHA regarding Interventional Therapies for Acute PE:
ahajournals.org/doi/10.1161/…
Presented simultaneously at @PERTConsortium PE symposium.
Summary in this thread.
I’m no mathematician but the below appears more scientifically rigorous and transparent (with hopefully more room for accompanying actionable knowledge development) than the well publicized recent Open AI math proof dump.
Perhaps this is what mathematicians are upset about?
🧬 Training and evaluating AI scientists on drug target discovery is hard.
We introduce DrugTargetWorld: Synthetic Biomedical Worlds for Training and Benchmarking AI Scientists
The environment turns end-to-end drug target discovery into a problem where AI scientists can be evaluated against causal truth.
🎯 We’ve been able to hill climb on problems like Chess, Go, StarCraft, and even now computer science and mathematics because they have a verifiable reward.
We want to bring that to drug discovery.
Built with @BGomes_1905 @euanashley and an incredible team at Stanford and beyond.
1
7
1,000
Great summary here!
safe to say that LPa reduction, if effective, is more incremental tx than statins, PCSK9i, GLP1a
But can’t disregard more potent agents with different mechanisms of action in perhaps better selected populations still in trials. Is LPa in its hype cycle?
Some take aways from what we know about HORIZONS phase 3 trial of Lp(a) lowering
1. We know very little. We only know the primary endpoint in the entire population.
2. We do not know if some subgroups may have benefited more like those with a higher baseline Lp(a).
3. Do those who were not on dual or single anti platelet therapy benefit?
4. Did efficacy vary by baseline and on treatment LDL?
5. Did efficacy vary by achieved Lp(a)?
6. We need more sophisticated analyses of the particle composition and oxidizing potential of the patient’s Lp(a) particle to assess for heterogeneity in response.
7. Were the results different in MI patients vs stroke vs PAD patients? These vascular beds behave differently in response to therapeutic interventions.
8. Were the results different in patients with single vessel disease vs multivessel disease where there were more targets for modification.
9. This was a time to first event analysis and not a time to any event analysis (a multiple event analysis). A multiple event analysis captures disease burden and may be more highly powered.
10. For those who were on dual antiplatelet therapies, once these were discontinued, was there an event reduction which would suggest optimal antiplatelet therapy may have reduced residual risk so much it could not be modified?
11. Much has been made of observational data suggesting aspirin may modify risk in patients with an elevated Lp(a). Was there an interaction with aspirin therapy ?
12. Mendelian randomization / genetics captures lifetime exposure and not several years of exposure as in this trial. Were the KM curves diverging late suggesting prolonged therapy might have been needed?
13. Were the assumptions regarding the magnitude of the potential efficacy signal too optimistic and was the trial underpowered?
14. In a trial this long patients may go missing. How good was the follow up? Was there informative censoring?
15. The results of primary prevention trials where LDL lowering and antiplatelet therapy may not have been as aggressive may still show benefit.
16. This ONE trial does not negate the potential causal role of Lp(a). The LDL and antiplatelet therapy may have reduced residual risk so much that there was very little risk to be modified. A trial Lp(a) lowering only would share more light on this question (a trial will never be done). What would happen if all patients were on Lp(a) lowering therapy. Would LDL lowering further improve outcomes? We don’t know for sure.
17. Was there an interaction with CRP, gender age risk score?
18. Obviously there are more questions than answers at this point. I raise these questions as exploratory hypotheses, to further our understanding of the biology, not to salvage a neutral trial.
We all anxiously await the full recitation of results and answers to these questions as well as others.
Thank you to the patients who participated.
12
1,544
This is a big one. Couple of related thoughts:
1) makes data from pcsk9i trials look even more impressive.
2) AI won’t solve drug discovery as easily as tech enthusiasts (and I am one) would hope. The vital outcome trial barrier is one that will not be overcome by an LLM.
Breaking and MASSIVE:
Lowering Lp(a) with #pelacarsen in patients with elevated Lp(a) and heart disease does NOT improve cardiovascular outcomes in Phase 3 trial globenewswire.com/news-relea…
The world’s premiere interventional cardiology journal could not be in better hands!
I am deeply grateful for the outpouring of support and kind messages following this week’s announcement.
My sincere thanks to the @ACCinTouch and @JACCJournals leadership for their trust, and to the colleagues, mentors, trainees, patients, friends, and family who have shaped this journey.
I am tremendously excited about what lies ahead. JACC CVINT has a remarkable legacy, and I look forward to working with the interventional community to build on that foundation and shape an ambitious next chapter together.
Thank you for your encouragement and partnership. I look forward to all that we can accomplish together.
@MayoClinicCV @MayoClinic
Monopolies breed inefficiency and extract ill-gotten profits. Regulatory bodies like @ABMSCert are no exception.
Happy to see Florida take this anti-monopolist action and hope other states follow its lead.
Important development in the national @ABMSCert MOC discussion: the Florida Board of Medicine has approved @nbpasadvocacy as a recognizing agency: "Once the final order is issued, physicians who hold qualifying NBPAS certification will be permitted to identify and advertise themselves as board certified in Florida."
flmedical.org/florida/Florid… @FloridaMedical
Nice explanation here. I can say from personal experience that, due to current federal policies, the US is already losing out on unique, world class talent that would pioneer novel research and deliver cutting edge clinical care to populations that need it.
Academic promotions are an obvious use case for AI.
A post trained LLM (or well prompted agent) could easily eval impact against internal criteria & peer institutions.
Beats current system of eval run by folks without domain specific expertise in both speed & objectivity.
Abolishing non-competes offers a market-driven alternative to the rising tide of physician unionization.
Would create true market competition.
We shouldn't settle for a commoditized workforce when we could have a system where docs thrive & continually strive for excellence.
Breaking news! One of Maryland's largest health systems has ended non-compete clauses in physician contracts. Kudos to @MedStarHealth for doing the right thing. Great work @ACCMaryland and @MedChiupdates for the legislative effort.
How about you @HopkinsMedicine @umms @LBHealth - how do you treat physicians?
@Cardiology @sxz110 @SrihariNaiduMD jacc.org/doi/10.1016/j.jacad…
Congrats to @MedStarHealth for showing confidence in their operational leadership by eliminating non-compete agreements!
Now, state legislatures must step up. Hospitals have a game-theoretic incentive to keep non-competes in place unless mandated otherwise by law.
🔥 Congrats to investigators from 73 global sites on completing enrollment in PEERLESS II !
Thank you to entire steering committee & @InariMedical for your unwavering support!
And thank you most of all to 1200 patients for trusting us with their care!
jscai.org/article/S2772-9303…
Always bittersweet when we wish congratulations to our graduating senior interventional/structural cardiology fellows @PennMedicine: @MonicaTung, Steph Colello, & Ayesha Khan!
We welcome the leadership of our rising senior fellows: @TayyabShahMD, @andrewjarrah, & Bob Geis.
Congratulations to all the investigators for this landmark achievement!
Huge admiration and gratitude for @akhileshsistaMD who has spent a decade of his life dedicated to accomplishing this.
And thanks most of all to the 500 participating patients.
So called “pay for performance” programs are just another facet of the medical metric industrial complex which over and over again shows itself to be a public health threat, despite being undergirded by seemingly good intentions.
@Amgen picks up an all star!
I'm very excited to share that I will be joining @Amgen, overseeing cardiometabolic research and human genetics, starting Aug 3!
I am excited to shape target identification, biology, and drug discovery as the therapeutic area head for cardiovascular disease, obesity, and general medicine. And I will lead the human genetics efforts, spanning Amgen deCODE genetics (a subsidiary of Amgen), that drive target discovery and biology across all therapeutic areas.
I have been incredibly fortunate to interface with truly remarkable individuals and culture across @MassGenBrigham, @broadinstitute, and @harvardmed, where I could not have asked for a more meaningful and enjoyable academic career.
Being able to build on my prior work, discover and clarify therapeutic hypotheses, and now carry them through to medicines that reach millions is an opportunity I'm grateful for. I'm very excited for what comes next.
3
1,055
For those attending #ACC26:
This will be a great review of the new @AHAScience / @ACCinTouch Acute PE Guidelines!
Join us at #ACC26 for a fast-paced, game show–style showdown where top FIT teams from across the country compete for the title &
prizes 💰
🔥 Interactive. Competitive. High-yield.
📍 La Nouvelle C
🗓 March 29, 2026
⏰ 13:45 – 14:45
Don’t just learn the guidelines—beat them!
I’m glad that experts making foundational advances in a field are included in the guideline writing process.
The AHA/ACC have strict policies to manage industry relationships (including majority & Chair with no relationships):
professional.heart.org/en/gu…
Guidelines should be written by Neutral Martian types WITHOUT financial dualities of interest.
THIS IS UNSEEMLY 👇🏻
4
5,132
Is there a conflict of interest when hospitals:
1) design policies to prevent pt “leakage”
2) incentivize intensity of care with RVU targets
3) use 340b arbitrage
There’s an honest convo to be had re: pros & cons of competition from MD-owned hospitals.
The below is propaganda.
There is no issue with physician-led hospitals- the issue is about the conflict of interest when physicians self-refer patients to their own hospitals.
The data is clear: POHs tend to treat more commercially insured and healthier patients than full-service hospitals.
In rural communities, this can leave rural hospitals with a greater financial burden, further threatening their ability to keep their doors open and keep 24/7 care available in their communities.
Read more: fah.org/wp-content/uploads/2…
This editorial allowed for the guideline chairs to explain why Vesalius was not included as evidence in the guideline while contextualizing the results of the trial within its reccs.
Is it useless to answer a common question about the document?
Not personal but I’m not sure of the wisdom of having guideline authors write editorials on their own guideline 👇🏻👇🏻
2
4
4,323