
founder scientist — cancer / ai | ex @ICR_London | PhD @BristolRobotLab
London, England
Joined April 2022
- Tweets3.9K
- Following229
- Followers271
- Likes2K
Khalid Abduljabbar retweeted
I am coming home to the UK in December! We will be hosting some @CompoundVC events the first week of December and encourage y'all to join!
-> 12/1 Compound Idea Factory in London - luma.com/o0pq0nw8
-> 12/2 Oxford Biotech Builders Lunch - luma.com/lkvcz7ux
-> 12/3 Cambridge BioxAI Breakfast - luma.com/whjmrsyq
Khalid Abduljabbar retweeted
So excited to share more about what I have been up to. @ProximalVC is a first check fund that loves to partner with founders before incorporation to support ideation and initial GTM strategy.
If you’re building the systems that make discovery smarter, trials faster, manufacturing more resilient, supply chains more transparent or commercial execution more precise - please reach out!
Former SignalFire investor Sahir Raoof raises $27.5M for new fund targeting AI in pharma
spr.ly/6014ByDyxC
Khalid Abduljabbar retweeted
One more week to complete our survey! We want to understand what clinicians need to bring ctDNA MRD testing into routine clinical practice. If you care for patients with breast cancer, please take a few minutes to participate and help shape future research!
@nlinmd @stolaney1
Dr. Ilana Schlam (@IlanaSchlam) discussing her new #ctDNA study at the HCC #BreastCancerResearchRetreat. The study is on the use of ctDNA NGS and MRD Assays in Clinical Practice. If you are a licensed clinician who treats patients with #BreastCancer, help us better understand current practice patterns and perceptions regarding the use of circulating tumor DNA (ctDNA) standard next-generation sequencing (#NGS) assays and minimal residual disease (#MRD) assays in breast cancer, across both early-stage and metastatic settings.
👉Study Link: redcap.link/ctdna_provider
Khalid Abduljabbar retweeted
1/3 Excited to share our review in @NatureRevCancer with @trevoragraham & @AndreaSottoriva!
Inspired by Dobzhansky, we believe cancer genomes make much more sense in the light of evolution, and so we ask ourselves: how can we read a tumour's evolutionary history from its genome?
Khalid Abduljabbar retweeted
Camizestrant is now approved for 1.5L treatment of patients with HR+ MBC and emergence of ESR1 mutations.
Will you start monitoring for ESR1 mutations with repeat ctDNA testing during 1L treatment?
21%Yes, every 2-3 months
19%Yes, but less frequently
28%No / Not feasible
32%See results
320 votes • Final results Khalid Abduljabbar retweeted
🆕Our Sept issue is now live!
📊Articles: #hepatocellularcarcinoma #NSCLC #rectalcancer #breastcancer #prostatecancer
📉Featuring trials: EMERALD-3, HARMONi, ALINA, STAR-TREC, & GETUG AFU 18
🎨Cover: breast cancer, TILs, and AI
🖱️Read here: thelancet.com/issue/S1470-20…
Khalid Abduljabbar retweeted
Today, featured on the cover of The Lancet Oncology, two publications add to the zero-shot validation of Case45's AI biomarkers.
Khalid Abduljabbar retweeted
Excited to share our new paper on AI and manual TILs in early HER2-positive breast cancer (NSABP B-41). AI-based TIL assessment was associated with pCR across ER subtypes and may complement manual TIL scoring
nature.com/articles/s41523-0…
@khalidated @SandraSwainMD @DFCI_BreastOnc
Khalid Abduljabbar retweeted
1/n Immensely proud of this one.
My lab's new paper in @NatureGenet — led by a physicist (@niccalonaci) and a clinician training in AI (@KrasniqiEriseld ) — just went live.
2 years of work that show how computational tools can help cancer medicine 🧵👇
🧬 Not all driver mutations are equal - whether the WT allele is lost, the mutant copy is gained, and how many copies carry the lesion matters.
Our new work is now out in Nature Genetics!
DOI: 10.1038/s41588-026-02666-z
@KrasniqiEriseld @gcaravagna
Khalid Abduljabbar retweeted
When we seeded Denali the idea was to break the curse of the blood/brain barrier. It took a decade and tons of faith and money. Biotech is hard. Curing diseases is hard. This is a good summary. markets.financialcontent.com…
Khalid Abduljabbar retweeted
Loving all the great London stories from the last few weeks 👇🏽
London’s been ranked a top place in the world to raise children.
Tech and AI companies are thriving in the capital.
Oxford Street will be pedestrianised.
A new open water swimming spot has been planned for the Thames. The list goes on.
London is the greatest city in the world.
Khalid Abduljabbar retweeted
⭐️Lung cancer in never smokers: from early detection to prevention- Nice review on this by @CharlesSwanton & colleagues @trendscancer @CellCellPress
👉Lung cancer in never smokers (LCINS) now accounts for a growing proportion of lung cancer cases, with distinct demographics and biology.
👉Identifying risk factors for LCINS remains challenging. The most well-established factors—germline variants, clonal hematopoiesis, and environmental exposures—currently lack validated screening tests @OncoAlert
cell.com/trends/cancer/fullt…
Khalid Abduljabbar retweeted
I lost my brother to cancer.
I was with him when he was told his diagnosis - a moment I will never forget. Throughout, the NHS respected and cared for him.
When he passed away, it hit me like a bus - even though I knew it was coming.
I'm determined that every person diagnosed with cancer gets the best possible chance of beating it.
Our National Cancer Plan is the most ambitious in a generation. It means earlier diagnosis, slashing waiting lists and investing in cutting-edge technology to build an NHS fit for the future.
Khalid Abduljabbar retweeted
GS1-05 – #SABCS25
Manual, digital, and AI-based TILs scoring all predict prognosis and pertuzumab benefit in HER2+ eBC in APHINITY:
- Manual scoring showed high reproducibility
- Higher TILs were consistently linked to lower recurrence risk (across all methods)
- AI spatial metrics strengthened prognostic accuracy
- High-TIL patients derived additional benefit from pertuzumab
@OncoAlert @SABCSSanAntonio
Khalid Abduljabbar retweeted
Amazing presentation by Roberto Salgado from @TILsWorkGroup
TIL assessment of APHINITY using CASE45 AI model and manual TILs (great ICC)
Combining AI and manual assessment is a better prognostic biomarker than either one of them
Khalid Abduljabbar retweeted
Very big news out today-- adjuvant abemaciclib improves overall survival in high-risk ER+ breast cancer at ~7 yrs of f/u! Showing OS benefit in this setting is not easy & nice to see what an incredible impact adj cdk4/6i can have for our patients!
prnewswire.com/news-releases…
Khalid Abduljabbar retweeted
For those interested in reading the full manuscript, here’s a Free Access Link: secure-web.cisco.com/1J1Rutx…
How do different breast cancer treatments perform immediately after T-DXd? In our latest #JNCI paper, we analyzed data from ~800 pts with MBC in Flatiron, finding promising rwPFS with several regimens, yet some degree cross-resistance between T-DXd and SG. academic.oup.com/jnci/advanc…
Khalid Abduljabbar retweeted
Our new preprint to scale single/cell differential expression testing with millions of cells from multiple individuals. A beautiful combination of statistics and implementation for advanced GPU architectures. Kudos to @giosantacatt and Niccolò for driving this beautiful work!
1/ Meet DEVIL, a Bayesian framework for single-cell DE testing that is patient-aware and GPU-powered. It delivers robust results across millions of cells in large cohorts, bridging the gap between statistical rigor and computational scale.
🔗biorxiv.org/content/10.1101/…










































