Geoff Macintyre

Checkout the latest preprint from the lab - single-cell CIN signatures unlocked! Ongoing HRD improves PARPi sensitivity detection and ongoing NHEJ associates with subclonal diversification in TNBC

Fluctuating DNA methylation tracks cancer evolution at clinical scale Nature - Cancer evolutionary dynamics are quantitatively inferred using a method, EVOFLUx, applied to fluctuating DNA methylation.

Cancer is an evolutionary disease, but does knowing a cancer’s evolutionary past help predict its future? Out today in @nature, we learnt the evolution of 2000 lymphoid cancers and found it was highly correlated with clinical outcomes! (1/7)
rdcu.be/eFrrc

13/ Want to know the genesis story of this research? Check out the “behind the paper” post communities.springernature.com/posts/toward...

12/ Thanks to patients and funders for their support @cniostopcancer.bsky.social @isciiisalud.bsky.social isciii.bsky.social @cienciagob.bsky.social #BecariosFLC #illumina @innovateuk.bsky.social @cuh.nhs.uk #TailorBio @cruk-ci.bsky.social

11/ Kudos to @jsneaththompson.bsky.social, Laura Madrid, @bhernando.bsky.social & co-authors for driving this work!

10/ What’s next? We’re funded by @mintradigital.bsky.social #NextGenerationEU for analytical validation and will be ready to run prospective trials in 2026. From organoids to algorithms to patients: precision chemo is possible!

9/ To ensure a flexible pathway to the clinic, we also tested biomarker reproducibility in ctDNA samples and TSO500 panel data, bringing us closer to real-world implementation

8/ It worked! We emulated trials to validate resistance predictions to platins, taxanes & anthracyclines across ovarian, breast, prostate & sarcoma

7/ Next step? A prospective trial? We tried but couldn’t! No one wanted to run/fund a trial using “old” chemos. So we had to get creative. Luckily, as chemos are widely used, there was a wealth of real-world data eg TCGA & @HartwigMedical to emulate biomarker trials - even RCTs!

6/ Our results looked great!

5/ Next step? Proof-of-concept using retrospective data from 50 ovarian cancer samples. Ovarian cancers were ideal as all 3 chemotherapies are routinely used. We focused on predicting resistance. Why resistance? Because it allows patients to avoid toxic side-effects

4/ We focused on optimising 3 CIN signature-based biomarkers to classify patients as resistant or sensitive to 3 commonly used chemotherapies: platins, taxanes or anthracyclines. Our goal: to optimise biomarker thresholds to use pan-cancer

3/ These prelim data showed correlations between CIN signatures and chemotherapy response. As the full spectrum of CINsigs can be quantified in a tumour using a single genomic test, we hypothesised that CINsigs could predict resistance to multiple chemotherapies at diagnosis

2/ Back then, we already had preliminary data suggesting these CIN signatures may be useful as therapy response biomarkers, mainly via synthetic lethality with the mechanism of action of the drug (CIN signature➡️defective pathway➡️dependency, which the drug exploits)

A pan-cancer compendium of chromosomal instability - Nature Copy number signatures characterize different types of chromosomal instability and predict drug response.

1/ 3 yrs ago we developed a computational framework to decode chromosomal instability www.nature.com/articles/s41...: input a tumour genome➡️output CIN signatures. As these CIN signatures represent different causes of DNA damage, they provide a read out of defective pathways in a tumour

🚨Chemo treatment upgrade!🚨

Check out our approach to modernise chemotherapy treatment published today in @natgenet.nature.com. From @cniostopcancer.bsky.social #TailorBio @cruk-ci.bsky.social www.nature.com/articles/s41... More details 👇

@blaschaves.bsky.social @mescobarrey.bsky.social @torresmarina.bsky.social

…and to the patients and funders @cniostopcancer.bsky.social ‪@isciiisalud.bsky.social‬ @cienciagob.bsky.social #BecariosFLC #H12O

We hope this study will inspire further forecasting efforts across other molecular alterations. Thanks to all members of the @gmaci.bsky.social Lab & collaborators Paz Ares Lab #PPCG 14/

TL;DR Forecasting oncogene amps & tumour suppressor dels is feasible! This can refine risk stratification and anticipate treatment resistance, paving the way for earlier, smarter and more personalised cancer care. There is much more in the preprint so check it out! 13/

MET amps cause EGFRi resistance in ~25% of NSCLCs. Forecasting MET amp in 33 EGFR-mutant NSCLC tumours treated with osimertinib showed high-risk patients had shorter PFS & OS. This can be used to flag candidates for upfront EGFR+MET inhibition (eg MARIPOSA trial) 12/

Currently, LGGs are classified into 4 WHO risk groups. CDK4/PDGFRA amps and CDKN2A dels are linked with poor prognosis but under utilised. Forecasting these facilitates a risk upgrade of 9% of IDHmut-non-codel cases while maintaining median survival times across WHO groups 11/

Encouraging right? We then applied our approach to two clinical scenarios where forecasting specific genetic changes might unlock new clinical opportunities: risk stratification of low-grade glioma (LGG) and anticipation of osimertinib resistance in lung cancer 10/

Next we tested longitudinal pairs forecasting at the early time point (before driver amp) and testing at the latter. In prostate, we predicted AR amp (linked to ADT resistance) in pretreatment samples. In NSCLC, we predicted HIST1H3B amp (exclusive to metastases) in primaries 9/

First, we tested performance on two independent cohorts: PCAWG: 2,114 primaries; HMF: 4,784 metastases. 147 of the 241 models showed AUC > 0.7 across both datasets 8/

We trained this model for 241 drivers using 7,880 TCGA samples across 33 tumour types. But do these models work? 7/

Challenge 3: forecasting in a clinical setting. Solution: binarise predictions and only use standard genomic test data as input. We designed guidelines to apply and (if needed) train the model + optimize thresholds for binary risk classification (high vs low) 6/

Challenge 2: growth rates of mutant vs non-mutant cells (and thus selection) cannot be easily determined in a clinical context. Solution: approximate selection coeffs using driver amp/del frequency at a population-level (supported by recent work showing s≈fβ) 5/

Challenge 1: amp/del rates cannot be readily measured from an input genome. Solution: approximate using a steady-state probability of locus-specific copy number change over tumour lifetime. We adapted our previous CIN signatures (CX bit.ly/42SVA3i) for this 4/

Simple model? Seemingly so. But estimating mutation rate and selection coefficients from tumour DNA alone, especially for DNA copy number, is tough! We faced a number of challenges: 3/