David Tamborero

Please read our @myESMO recommendations on how to report NGS results of solid tumors to clinical teams - it was great to work on this, fantastic effort!  x.com/quimmateo/stat…

New @Nature study: >50% of lung cancer metastases are seeded by other metastases, not the primary tumor. This "seeding from seeding" reveals a complex evolutionary cascade that allows cancer to colonize the body. nature.com/articles/s4158…

🚨⭐️Effect of one or two cycles of dual immunotherapy with nivolumab and ipilimumab in patients with mismatch repair-deficient rectal cancer (RESET-R): interim results from a multicentre, single-arm, phase 2 trial - The Lancet Gastroenterology & Hepatology thelancet.com/journals/langa…

New @CellCellPress review by Augusto Villanueva Director of the @Perlmutter_CC Liver Tumor Program: Hallmarks of liver cancer: Therapeutic implications cell.com/cell/fulltext/… @nyulangone @NYULH_GastroHep

Recommended📚: Multi-layered molecular profiling informs the diagnosis and targeted therapy of desmoplastic small round cell tumor, with @StefanFrohling @DKFZ @NCT_HD as Senior Corresponding author, now published in @NatureComms 🔗 nature.com/articles/s4146…

🚨 Not all HER2+ tumors respond equally to dual blockade… 👉 Tumor genomics may be the missing piece. A new analysis of HER2+ early breast cancer shows: 🧬 PIK3CA mutations = lower pCR •47.7% vs 66.7% with dual HER2 blockade •OR ~0.4 → clear resistance signal 💉 Striking finding with chemo backbone •Nab-paclitaxel: ❌ 38.7% vs 72.0% pCR •Paclitaxel: ➖ No major difference 👉 Suggests interaction between biology + chemo choice 🧠 Key insight •Resistance seen ONLY with anti-HER2 therapy •NOT seen in chemo-only cohort 📊 Translation: 👉 PIK3CA ≠ chemo resistance 👉 PIK3CA = HER2-targeted therapy resistance ⚠️ TP53, MAPK mutations → NO impact on pCR 💡 Clinical takeaway •Should we start genotype-guided neoadjuvant strategy? •PI3K pathway targeting may be critical in HER2+ 🔖 This will change how we think about “non-responders” ❓ Do you check PIK3CA upfront in HER2+ disease before neoadjuvant therapy? 📖 Full paper in comment ⬇️ #OncoTwitter #MedTwitter #BreastCancer #HER2 @OncoAlert @myesmo @esmo_open @ASCO

A new preprint reports that genetic loss of JAK1 increases susceptibility to HPV infection and subsequent non-melanoma skin cancer (NMSC) risk. This human genetic evidence arrives at a critical moment. Selective JAK1 inhibitors like upadacitinib and abrocitinib are a major drug class in modern medicine, approved for atopic dermatitis, rheumatoid arthritis, psoriatic arthritis, ulcerative colitis, and Crohn's disease. Millions of patients, many of them young, take these drugs for years or decades. The new genetic data hints HPV susceptibility and NMSC risk as potential long-term adverse effects of these drugs. The story gets more interesting when you look at how JAK inhibitors evolved. Early versions like tofacitinib were non-selective; they are effective, but caused hematological toxicities (from JAK2 inhibition), and lymphocyte depletion (from JAK3 inhibition). Researchers eventually recognized a useful dissociation: the therapeutic benefit in autoimmune disease flows primarily through JAK1-dependent pathways (IL-6, IL-4, IL-13 signaling), while the most visible toxicities came from the other isoforms. That insight drove the field toward JAK1-selective agents, which are now the dominant class. The logic was sound for the toxicities the field could measure. What nobody systematically evaluated was JAK1's specific role in antiviral skin immunity. JAK1 GoF causing autoimmunity was well characterized, but JAK1 LoF in humans wasn't, except for one patient with biallelic mutations and broad immunodeficiency. Interestingly this patient also presented with warts, which didn't stand out at that time. There were logical reasons to worry about interferon signaling and viral susceptibility in JAK1 inhibition, but no human genetic evidence was there to back that concern. Fan et al. (medRxiv, April 2026) now fill that gap. Across four independent pedigrees with epidermodysplasia verruciformis (EV), a Mendelian disorder linking HPV to skin cancer, they identify heterozygous JAK1 loss-of-function variants as causative, establishing that JAK1 haploinsufficiency permits persistent β-HPV infection and NMSC development. Scattered case reports of warts and NMSC in patients on JAK1 inhibitors exist in the literature, and NMSC signals seem to exist in trial data, though without HPV genotyping info. These signals weren't strong enough to motivate anyone to look harder. Fan et al. data now will motivate drug developers to retrospectively analyze their trial data to quantify this important adverse outcome. This is another beautiful example of rare disease patients informing on possible adverse effects of drugs used for common diseases. Fan et al. medRxiv 2026 medrxiv.org/content/10.648…

🧬 #ctDNA and the preclinical window: are we closer to true early detection? acsjournals.onlinelibrary.wiley.com/doi/10.1002/cn… @OncoAlert New data begin to quantify something we’ve long sensed, but never clearly measured: 👉 how long tumors are actually detectable by ctDNA before clinical diagnosis 🔹 1. A narrow but actionable window • Median ctDNA detectability: ~0.7–1.2 years depending on stage • Highly variable across tumors (e.g., pancreas shorter, lymphoma longer) 👉 This is not a wide window, but it may be enough 🔹 2. Biology matters more than technology • ctDNA positivity reflects biologically aggressive disease • HR for mortality ~2 when ctDNA-positive vs negative 👉 We are not just detecting cancer, we are detecting clinically meaningful cancer 🔹 3. Implications for screening strategies • Data support ~annual screening intervals • Reinforces feasibility of multicancer early detection (MCED) approaches 👉 Timing may be as critical as sensitivity 💡 My take: This study reframes ctDNA from a biomarker → to a temporal signal of cancer evolution Not all tumors give us time. But for those that do… this may be the window where outcomes can truly change. @ecancer @MedicalwatchHQ @urotoday #ctDNA #LiquidBiopsy #EarlyDetection #MCED #PrecisionOncology #OncoTwitter

Don’t miss our editorial on Lancet Oncology about the Ideate-PanTumor trial (with an anti-B7H3 ADC). Could it potentially become a treatment option for SCLC? A work with Jean Marie Michot doi.org/10.1016/S1470-…

🧬 Introducing: Knockout Discovery Engine Some people are born with naturally broken genes. People with a broken CCR5 gene are immune to HIV. People with a broken HSD17B13 gene have lower rates of chronic liver disease. Drugs backed by this kind of genetic evidence are 2.6x more likely to get approved. I built this to automate the search for such genes 🧬 How it works: → type drug description → analyze human genetic data → get drug efficacy evidence report Demo ↓

KRASG12C inhibition meets EGFR feedback Block both KROCUS trial. Fulzerasib + cetuximab in 1L NSCLC (phase 1b/2) in @TheLancetOncol ORR 69%, PFS 12.5 mo Rational combination. Promising early signal thelancet.com/journals/lanon… @OncoAlert

HER2 heterogeneous breast cancer models reveal novel therapeutic targets and subclonal dynamics during evolution to resistance to HER2-targeted therapies ~40% of HER2+ tumors are heterogeneous. In preclinical models, T-DXd achieved complete regression in homogeneous tumors but not in heterogeneous ones. HER2 low subclones are less sensitive to ADCs, persist under treatment pressure and drive residual disease, while remaining sensitive to HER2 TKIs. It’s clonal selection. Treating only the dominant HER2-high population isn’t enough. The minor clone shapes what comes next. Combination strategies aren’t optional. aacrjournals.org/cancerdiscover… @OncoAlert

Online Now: Cancer type-specific variation in patterns of driver alterations across 50,000 tumors dlvr.it/TRk4T8

How EGFR inhibitors transformed outcomes in EGFR-mutated non–small-cell lung cancer

MSK-IMPACT is truly the gift that keeps on giving! Cancer type-specific variation in patterns of driver alterations across 50,000 tumors (!!) cell.com/cancer-cell/fu… "One-third of all drivers arose in non-canonical contexts" Congratulations @MFBerger1 @DSolit & team.

⭐️Wow 👉🏼published in @Nature 👉🏼“Thymic health and immunotherapy outcomes in patients with cancer.” ⭐️Across melanoma, breast, & renal cancers, the signal is clear: thymic health is a pan‑cancer, #tumoragnostic determinant of immunotherapy efficacy. ⭐️I think there are huge implications for patient stratification, treatment timing, & immune‑rejuvenating strategies in #PrecisionMedicine here @OncoAlert nature.com/articles/s4158…

MTAP Loss Is Frequent in Oncogene-Driven NSCLC and May Confer Sensitivity to Combined PRMT5 Inhibitors and Targeted Therapies In >13,000 tumors, MTAP deletion was found in: ALK 27–45% RET 18–35% EGFR 17–29% 98% co-deleted with CDKN2A. TKI outcomes unchanged ⚠️⚠️ annalsofoncology.org/article/S0923-… #NSCLC #lcsm #EGFR

A fascinating discovery of a set of mosaic mutations in a pair of monozygotic twins reveals a new mutation phenomenon. During genetic investigation for suspected neurodevelopmental disorder (NDD), the authors identified two mosaic loss-of-function mutations (1 bp deletion causing frameshift) 3 bp apart in HNRNPU, encoding ribonucleoprotein. Mutations in HNRNPU cause autosomal dominant NDD. Both mutations sit on the paternal allele, but they never occurred together in the same cell. Each mutation is seen in distinct cell populations. At first glance, one could wonder if these somatic mutations might have arisen from two independent mutation events. However, the likelihood of two such mutation events occurring simultaneously in close proximity on the same allele but in different cells during early development is vanishingly small. The authors propose an alternative mechanism. A single double-stranded DNA damage event may have occurred in an early embryonic cell. During repair, the two sister chromatids were processed slightly differently, generating two nearby but distinct deletions. When the cell divided, the chromatids segregated into different daughter cells, producing a mosaic individual with two distinct mutant alleles that never coexist within the same cell. The authors call this configuration clustered monoallelic mosaicism (cMoMa). The proposed mechanism challenges a common assumption that one mutation event leads to one variant. The findings suggest a mechanism where one mutation event could lead to multiple lineage-segregated variants. If this mechanism is real, it could represent a previously underappreciated path of human mutagenesis. Böhnlein, Maass, Dennig, et al. medRxiv 2026 medrxiv.org/content/10.648…

When a phase 1 trial is published in @NEJM, you can bet it will be impactful. Here, the p53 reactivator rezatapopt showed an ORR 20% among 77 pts with TP53 mutant (Y220C) advanced tumors. Are we getting closer to drug the most undruggable of all mutations? nejm.org/doi/full/10.10…

New paper alert📢! cell.com/cell/fulltext/… (@CellCellPress) - How does immune responses to solid tumors vary by age and tumor characteristics in children?! 🧵1/8

Top universities for Medicine, 2025. 1. 🇺🇸 Harvard University 2. 🇬🇧 University of Oxford 3. 🇺🇸 Stanford University 4. 🇺🇸 Johns Hopkins University 5. 🇬🇧 University of Cambridge 6. 🇺🇸 University of California, San Francisco (UCSF) 7. 🇬🇧 Imperial College London 8. 🇬🇧 UCL 9. 🇸🇪 Karolinska Institute 10. 🇺🇸 Yale University 11. 🇺🇸 MIT 12. 🇺🇸 UCLA 13. 🇺🇸 University of Pennsylvania 14. 🇨🇦 University of Toronto 15. 🇬🇧 King’s College London 16. 🇺🇸 Duke University 17. 🇺🇸 Columbia University 18. 🇸🇬 NUS 19. 🇺🇸 University of Michigan–Ann Arbor 20. 🇬🇧 University of Edinburgh (QS – Medicine)