iiinova — where your digital self keeps evolving

Barbara Hernando @bhernando.bsky.social · 12/09/2026

17/ Enormous effort from co-first authors Andreas Gruber (Konstanz), André Olsen (BRIC Copenhagen) and Kevin Cheng (OICR), and from @gmaci.bsky.social, @reimand.bsky.social and @jweischenfeldt.bsky.social who co-led it. Thanks to #PPCG, to the men who donated samples, and to our funders.

113

Barbara Hernando @bhernando.bsky.social · 12/09/2026

15/ Two more: non-canonical impaired HR was enriched in patients of African ancestry, and late-onset disease was dominated by replication stress while early-onset was enriched for HRD. Age at onset and aetiology are linked.

100

Barbara Hernando @bhernando.bsky.social · 12/09/2026

14/ Extending to mCRPC (HMF, n=240) we emulated an RCT: IMF6-high patients had markedly lower risk of treatment failure on ARPIs vs taxanes (HR 0.21). Retrospective, n=25 vs 94, hypothesis-generating – but mechanistically coherent via SPOP.

100

Barbara Hernando @bhernando.bsky.social · 12/09/2026

13/ Notably, the same Cox models on individual signatures, one class at a time, showed no strong risk associations in this cohort. The prognostic signal lives in the integration, not in any single class.

100

Barbara Hernando @bhernando.bsky.social · 12/09/2026

12/ Clinical: 4 IMFs (ROS, AR, non-canonical HR, canonical HR) carried HRs of 4–7 for metastasis vs non-CIN, independent of age, Gleason grade group, stage, TMB and late-to-early RT ratio. And they stratified WITHIN GG2/GG3 – the actual grey zone.

100

Barbara Hernando @bhernando.bsky.social · 12/09/2026

11/ Prevalence, in CIN tumours: AR-mediated mutagenesis dominant in 39%, replication stress 28%, ROS 21%. Replication stress being this pervasive in treatment-naive primary disease was not expected in a cancer we call slow.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

10/ Topography mattered. SNVs and SVs are enriched at tumour-specific ARBS — and SBS1 is DEPLETED there. Why? AR binds unmethylated CpGs. We confirmed it directly with Nanopore methylation in 8 tumours.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

9/ The eight: MMRd; ROS-mediated mutagenesis; AR-mediated mutagenesis; mitotic defects + replication stress; non-canonical impaired HR; replication stress; APOBEC; canonical impaired HR.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

8/ Challenge 3: every cluster had to earn its aetiology. Orthogonal evidence: driver status across 1,747 curated genes, DDR pathway deficiency, HRDetect, MSIsensor-pro, replication timing, hypoxia and AR pathway scores from RNA. Then replication in TCGA-PRAD.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

7/ Result: 8 integrated mutational footprints (IMFs). Important: these are NOT subtypes. Every tumour carries activities across several IMFs; what differs between patients is which processes are running and how strongly. A spectrum, not eight boxes.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

6/ Challenge 2: the four alteration classes aren't on a common scale, and mutational processes nest. Solution: normalise activities, then hierarchical clustering – swept parameters, arbitrated on stability, silhouette AND known biology.

120

Barbara Hernando @bhernando.bsky.social · 12/09/2026

5/ Full de novo extraction gave 22 SBS, 10 ID, 8 CX and 6 cSV signatures: a 46-dimensional description of each tumour. Two artefactual SBS excluded.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

4/ Challenge 1: the SV description was too coarse for prostate cancer. Solution: extend it. We classified each rearrangement by class × size × replication timing × fragile site – and, new here, by whether it fell at an AR binding site (ARBS). 6 stable cSV signatures came out.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

3/ Substrate: 959 primary tumours from 1,001 PPCG donors, 7 countries, uniformly reprocessed, germline + somatic called the same way, median 7 yrs clinical follow-up. 4.7M SNVs, 292K indels, 74K SVs, 55K CNAs.

110

Barbara Hernando @bhernando.bsky.social · 12/09/2026

2/ Motivation: signature analysis has been siloed by variant class. But one broken mechanism doesn't confine itself to one class — the same failing machinery that miscopies a base also duplicates a segment and shatters a chromosome. Read them separately, see four things.

100

Barbara Hernando @bhernando.bsky.social · 12/09/2026

1/ NEW PAPER 🎉 🧵 We mapped the mutational processes shaping 959 prostate cancer genomes from the Pan Prostate Cancer Group (PPCG), integrating SBS, ID, CX and complex SV signatures. Eight footprints account for the processes in 85% of cases www.nature.com/articles/s41... @nature.com

nature.com

Integrated signatures define mutational processes in prostate cancer - Nature

Eight integrated mutational footprints collectively explain the mutational processes in 85% of primary prostate cancer genomes.

154

Barbara Hernando @bhernando.bsky.social · 10/08/2026

Check it now! 💫 Exited to share this preprint, the result of a fantastic collaboration between Sarah Aitken’s and Martin Taylor’s labs. During my secondment at @yalecancer.bsky.social @yalepathology.bsky.social, I contributed to explore how H&E images can be used to predict tumour genomics

043

Reposted by Barbara Hernando

Sarah Aitken @s-j-aitken.bsky.social · 27/07/2026

🧬NEW PAPER🧬 To what extent is cancer development deterministic? Does the germline genome affect that predictability? Find out in our #StrainDifferences paper @nature.com "Genetic background sets the trajectory of experimental cancer evolution" www.nature.com/articles/s41... 🧵[1/14]

nature.com

Genetic background sets the trajectory of experimental cancer evolution - Nature

Experimentally replaying tumour evolution in divergent mouse strains reveals the importance of interactions between genetic ancestry and acquired cancer-driving mutations in shaping the earliest stage...

14717

Barbara Hernando @bhernando.bsky.social · 19/12/2025

Thanks for everything! I will miss being around in the lab but see you online 🫶🏻

021

Barbara Hernando @bhernando.bsky.social · 04/12/2025

Had a lovely dinner today with members of @s-j-aitken.bsky.social’s lab to celebrate my recent preprint! Sharing these small milestones (and the joy that comes with them) with friends is just magic ✨

130

Barbara Hernando @bhernando.bsky.social · 28/11/2025

This project has been with me since day 1 in the lab - a long journey of challenges, growth & incredible teamwork 🚀 So proud of @gmaci.bsky.social team for pushing this work to the finish line! Thks to collaborators (#Barbacid & #FSanchezVega labs) and @caixaresearch.bsky.social for their support

021

Barbara Hernando @bhernando.bsky.social · 30/10/2025

It was an honor to take part in the @yalepathology.bsky.social research retreat today! Great science, inspiring discussions, a fun “guess who?”, and wonderful people. Huge thanks to organizers for such a great work! Can’t believe only 2 months left to enjoy this amazing #AitkenLab team ✨

011

Reposted by Barbara Hernando

Sarah Aitken @s-j-aitken.bsky.social · 10/10/2025

A trio of talks this week: Grand Rounds @jhu.edu yesterday. Then @bhernando.bsky.social (who has only been here a fortnight!) and I gave talks at the NYC Genome Integrity Discussion group @rockefeller.edu. Thanks for inviting us Agata Smogorzewska and Philipp Oberdoerffer @yalepathology.bsky.social

092

Barbara Hernando @bhernando.bsky.social · 29/09/2025

Exciting 3 months ahead at the #AitkenLab @yaleschoolofmed.bsky.social! 🤩 Huge thanks to @s-j-aitken.bsky.social for the warm welcome, and to @caixaresearch.bsky.social for supporting my research career at all levels ✨

051

Reposted by Barbara Hernando

cniostopcancer.bsky.social @cniostopcancer.bsky.social · 23/06/2025

#CNIOStopCancer develop test that predicts which patients will not respond to cancer chemotherapy. They have identified biomarkers which, in clinical practice, would allow for more effective treatments and the avoidance of side effects. bit.ly/465QqnR

111

Reposted by Barbara Hernando

cniostopcancer.bsky.social @cniostopcancer.bsky.social · 23/06/2025

Un estudio del #CNIOStopCancer descubre biomarcadores que predicen qué pacientes no responderán a la quimioterapia contra el cáncer. Podrían servir para descartar tratamientos que no van a funcionar, pero provocan efectos secundarios y son caros. bit.ly/4lj1QJn

111

Reposted by Barbara Hernando

Geoff Macintyre @gmaci.bsky.social · 23/06/2025

🚨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 👇

nature.com

1179

Barbara Hernando @bhernando.bsky.social · 10/06/2025

💥New preprint from the @gmaci.bsky.social lab! Thrilled to share this major team effort, with #Angel as co-first author, now live on bioRxiv! Check out our threat to learn more about our work👇

010

Barbara Hernando @bhernando.bsky.social · 14/05/2025

Flying back from Bologna after the Pan Prostate Cancer Group #PPCG meeting. 3 days of fruitful discussions with lovely colleagues at the best venue I have ever been 🤩

030

Barbara Hernando @bhernando.bsky.social · 27/04/2025

Day 1 at #AACR2025 from my side! Meanwhile, at the #EMBO workshop on Chromosome Segregation and Aneuploidy, Maria Escobar will be presenting our project on ongoing CIN in KRAS mutant pancreatic organoids! A must meeting for the #CIN community meetings.embo.org/event/25-ane...

meetings.embo.org

Chromosome Segregation and Aneuploidy

Aneuploidy is a hallmark of cancer and developmental disorders such as Down syndrome. Understanding how cells accomplish faithful chromosome segregation and how chromosomal instability (CIN) impacts …

031

Barbara Hernando @bhernando.bsky.social · 26/04/2025

Ready for #AACR25! Excited for five full days of talks, posters and educational sessions in cancer research, and also for catching up with colleagues and friends 🤩

080