Thoracic Surgeon @RoswellPark Cancer Center. Cardiothoracic surgery @harvardmed, resident @WCMSurgery, former Surgeon @usairforce, AI researcher and proud Dad
While the end user of OpenEvidence is a clinician, everything we do is in service of providing those providers and, subsequently, their patients, with the resources they need to make informed decisions. Today, we take a big step into incorporating more resources beyond clinical evidence to support those decisions with Access Intelligence.
A prescription the patient can't afford, or the payer won't approve, doesn't treat anything.
Today we launched Access Intelligence: copay savings, patient assistance, program enrollment, and prior authorization, inside OpenEvidence. Free for all clinicians.
The help exists. There are thousands of support programs, and prior auths can be won. But the work is scattered across payers and programs, each with its own forms and criteria, and you have a few minutes per patient.
What's live now:
* Ask about availability and eligibility across 2,500+ patient support programs, updated daily.
* Send copay savings and discount offers to the patient with Patient TakeHome.
* Download pre-populated enrollment forms, or enroll on OpenEvidence.
* Write prior auth requests, appeals, and letters of medical necessity grounded in clinical guidelines, with PA checklists and relevant ICD-10 codes.
Choosing the right treatment should be the hardest part of getting a patient treated.
Recently published in #JTCVS Techniques: Complex reoperative tracheobronchoplasty for persistent, severe excessive central airway collapse with mesh erosion. Read more form surgeons at @BIDMCSurgery: hubs.la/Q04CVMjZ0@heidel_justin@SidhuGang
Wow! If such high pCR rates are replicated in larger trials, we next need to identify reliable algorithms to predict pCR and determine whether there are patients in whom we can omit surgery.
Can SABR replace surgery for operable stage I NSCLC?
New @JCO_ASCO study by Kagenaar et al. uses England's national cancer data (2015–2020) to compare outcomes in 12,546 patients:
🔹 Surgery: 10,247
🔹 SABR: 2,299
Study: ascopubs.org/doi/full/10.120…
🧵
I’d like to invite you to the Stronger Together Health Fair this Saturday!!! Our amazing ALCSI (@AmLungCSI) & GEDI (@gedihq) teams are hosting this event at Roxbury Community College from 10-1pm!! Everyone is welcome!! There will be lunch, free massages, and line dancing too!!
SANO trial #ESMO26
Why were distant mets more frequent with active surveillance (46% vs 29%) despite no significant OS difference?
▪️More intensive surveillance may have increased detection
▪️Locoregional regrowth during active surveillance may contribute to distant mets
▪️Higher non-cancer mortality after surgery may partly explain the lack of a significant OS difference
👉Perioperative chemotherapy, including D-FLOT, is now preferred for many patients with adenocarcinoma (~75% of SANO). In the SCC subset (~25%), DFS KM curves appear similar, but adjusted HR was 2.12 (95% CI, 0.68-6.61).
Careful discussion and shared decision-making are essential when considering NOM after nCRT for esophageal cancer.
🔗doi.org/10.1016/j.annonc.202…@Annals_Oncology@myESMO@OncoAlert
When we presented data from all the neoadjuvant/periop IO studies people were understandably concerned about what happened to those who didn’t get operated. People were also confused by the stage overlap between PACIFIC and the surgical trials. So the goal of MDT-BRIDGE was to better understand these gaps and evaluate whether an early MDT reassessment after 2 cycles could allow most patients to achieve definitive locoregional control, which they did in 95% of cases. MDT reassessment also enabled very high R0 rates. Complication profile for the CRT group seemed reasonable as well which was another gap in knowledge. The goal wasn’t AT ALL to compare the survival outcomes between the two forms of locoregional therapy. Our intent was to bring data to areas of uncertainty after multiple phase 3 trials. We also wished to understand why MDTs might interpret resectability differently and this work is still ongoing.
The data are also not meant to be used to justify a trial. Before such an RCT could be done, I think the Rad Onc community would have to prove that neoadj chemoIO followed by CRT is superior or at least equivalent to PACIFIC. Once this is done, we can talk about the RCT… still much work to be done.
Just published in Thoracic Surgery Clinics: a review of where AI and machine learning may actually improve risk assessment and complication prediction in thoracic surgery, and what it will take to move these tools into practice.
Free access: authors.elsevier.com/a/1ny3y…
Our clinical trial of AMIE, a patient-facing AI system, was just published in The Lancet.
I want to explain what *I* think this study ACTUALLY means.
It's about a fundamental question: how do we know that clinical AI works?
🧵⬇️
Our medical research system, AMIE, is the first patient-facing conversational diagnostic tool of its kind to be studied prospectively in a real-world clinical setting.
In a study, published today in @TheLancet, we found that when patients chatted with AMIE before an in-person appointment, the interaction built their confidence and helped them organize their thoughts. And physicians who reviewed the information from the patient conversation got more time back to focus on collaborative care and shared decision-making instead of digging through data.
Dr. Pat Seastedt (@SeastedtMD) gave a thought-provoking talk at #ITSOS2026 on where AI could change thoracic care: risk-based screening eligibility; NLP to find eligible patients missed by structured EHR data; AI-assisted CT interpretation, which may reduce missed referrals; more consistent preoperative risk assessment; and potentially measurable surgical quality through video analysis and process auditing.
His caution was clear: AI can make care more consistent without making it more equitable. We need to measure who benefits at every step, validate subgroup performance locally before and after deployment, and have thoracic surgeons help lead implementation.
Really impressed by this session at #ITSOS2026: “Disparities in Access and Treatment for Personalized Cancer Care.”
Every presentation was thought-provoking & each gave me something practical to take home and act on. Such an amazing line up!!!
Speakers:
Alex Potter, Harvard Medical School (@alexandra_p_24)
Amie Miller, Sarasota Memorial Health)
Fatima Wilder, MGB (@FWilder9)
Alexis Chidi, MSK (@AlexisChidi)
Patrick Seastedt, Roswell Park (@SeastedtMD)
Discussants
Loretta Erhunmwunsee, City of Hope (@LorettaEMD)
John Waters, UT Southwestern)
Moderated by Neel Chudgar, Montefiore (@nchudgar) and Jane Yanagawa, UCLA