Amit Kulkarni

#FDA recently approved #pembrolizumab- a tissue agnostic indication in metastatic tumors with high #TMB (>10 mut/MB) using Foundation One Dx companion diagnostic test. I want to share a tweetorial on why this indication could be problematic and hope to share some nuances on #TMB

@BrunaPellini Congratulations

✨Grateful to step into the role of Chief of Thoracic Medical Oncology at Baptist Health Miami Cancer Institute. Thank you to #DrLeonardKalman and #ManmeetAhluwalia for the trust and opportunity. Looking forward to building and growing a strong, impactful Thoracic Oncology program. @BaptistHealthSF @MiamiCancerInst baptisthealth.net/newsroom/press…

This paper from Harvard and MIT quietly answers the most important AI question nobody benchmarks properly: Can LLMs actually discover science, or are they just good at talking about it? The paper is called “Evaluating Large Language Models in Scientific Discovery”, and instead of asking models trivia questions, it tests something much harder: Can models form hypotheses, design experiments, interpret results, and update beliefs like real scientists? Here’s what the authors did differently 👇 • They evaluate LLMs across the full discovery loop hypothesis → experiment → observation → revision • Tasks span biology, chemistry, and physics, not toy puzzles • Models must work with incomplete data, noisy results, and false leads • Success is measured by scientific progress, not fluency or confidence What they found is sobering. LLMs are decent at suggesting hypotheses, but brittle at everything that follows. ✓ They overfit to surface patterns ✓ They struggle to abandon bad hypotheses even when evidence contradicts them ✓ They confuse correlation for causation ✓ They hallucinate explanations when experiments fail ✓ They optimize for plausibility, not truth Most striking result: `High benchmark scores do not correlate with scientific discovery ability.` Some top models that dominate standard reasoning tests completely fail when forced to run iterative experiments and update theories. Why this matters: Real science is not one-shot reasoning. It’s feedback, failure, revision, and restraint. LLMs today: • Talk like scientists • Write like scientists • But don’t think like scientists yet The paper’s core takeaway: Scientific intelligence is not language intelligence. It requires memory, hypothesis tracking, causal reasoning, and the ability to say “I was wrong.” Until models can reliably do that, claims about “AI scientists” are mostly premature. This paper doesn’t hype AI. It defines the gap we still need to close. And that’s exactly why it’s important.

Andexanet alfa pulled from the US market. This drug, that cost at ~$20k a dose, never showed real benefit, and caused stroke. The story is a fascinating example of misguided ethical thinking, and the human weakness for the power of narrative. This is a wild ride, get ready. 🧵

City of Hope welcomes Dr. Christine M. Lovly as head of our national thoracic oncology program. A leader in lung cancer research and care, Dr. Lovly brings deep expertise in precision medicine to help improve outcomes for patients nationwide. More: bit.ly/480Shea

A magical evening celebrating @cityofhopeoc. Huge congrats to our fearless leaders @AWalkerHealth & @DrEdKim, as well as our visionary CEO Robert Stone, who have guided us through a remarkable expansion. I recall when @franksstillhere brought up the notion of @cityofhope opening doors in #OrangeCounty - what seemed like a pipe dream over a decade & a half ago has now materialized. As the inimitable @DianaRoss sang for us tonight, “Ain’t no mountain high enough” … when it comes to #CityOfHope.

Sucralose consumption ablates cancer immunotherapy response through microbiome disruption ⁦@CD_AACR⁩ aacrjournals.org/cancerdiscover…

1/ We are thrilled to share our latest research in Cancer Discovery! Grateful to all collaborators in this work, work led by Paul Mischel and Ben Werner, and to @drimrannoorani and @magnushaughey for driving this work! Special thanks: team eDyNAmic, UCL, @TheCrick , and CGC/CRUK

@Joshua_Reuss @IASLC Congrats, Josh

Honored to present our data on neoadjuvant nivolumab and nivolumab/ipilimumab in potentially resectable diffuse pleural mesothelioma (DPM) at #WCLC25. @IASLC We evaluated perioperative nivo and nivo/ipi in epithelioid or biphasic DPM deemed resectable by multi-D assessment. 1/7

@chulkimMD Thanks Chul

@AmitKulkarniMD Congratulations Amit!

1/14 #WCLC25 Thrilled to be presenting a poster 📍(P1.11.71) “Preliminary Safety and Efficacy of Restorative Microbiota Therapy (RMT) with Chemoimmunotherapy in Metastatic NSCLC” on 📅September 7, ⏰10:30 AM to 12 PM. Brief #tweetorial here:

@shilpaonc Thank you!

@AmitKulkarniMD Congrats Amit! So proud of you!

To my friends at #WCLC25: please check out this incredible poster from @CityofHopeChi faculty member @AmitKulkarniMD. Amit joined us from @UMNCancer where he had cultivated an interest in #microbiome work & specifically fecal microbiome transplant (#FMT) . He wrote & received funding for this trial, which is opening soon at @cityofhope thanks to support from @AlanBryce9 & other colleagues. The exploration of chemoIO w FMT in #NSCLC is novel & early results appear promising.

@SMImmunology @cityofhope @DrRaviSalgia @AlanBryce9 @montypal @JyotiMalhotraMD Thank you, would love to collaborate if possible

@AmitKulkarniMD @cityofhope @DrRaviSalgia @AlanBryce9 @montypal @JyotiMalhotraMD Congratulations. We are planning a large randomized FMT + chemoIO trial in NSCLC with 180 patients.

On a recent work trip to the UK 🇬🇧, met with my aunt, @drkalyanipal (my dad’s little sis). She worked at @GSTTnhs & @ICR_London. I knew she spent a considerable amount of time in the lab, but had no idea she did early foundational work linking polycyclic aromatic hydrocarbons (#PAHs) & metabolites to chromosomal aberrations & downstream carcinogenesis. Have always been inspired by amazing #female figures in #cancerresearch but did not realize I was related to one! @ASCO @AACR @myESMO @cityofhope

Also published in @NEJM #MARIPOSA trial reassuring to see similar HR 0.75 MARIPOSA VS FLAURA2 - Overall similar results, different side effects, and $$ . Will access and patient preference determine selection of regimen ?

#WCLC25 Presidential symposium Dr. Planchard presenting #FLAURA2 OS data, HR =0.77, 12% absolute benefit at 3 yrs, with 72% of patients in Osi arm getting subsequent platinum chemo 🙌 ➡️longer period of osi treatment duration in combo arm ➡️Median time on chemo ~8 months

14/14 Lastly, I am very grateful to my current institution @cityofhope, and the leadership @DrRaviSalgia @AlanBryce9 for supporting to opening the study at Phoenix @cityofhope, hopefully very soon !!

13/14 Grateful to my mentors at @UMNews, Dr. Manish Patel and Dr. Robert Krazke, for their unwavering support. Thanks to AstraZeneca for supporting the study