Lightnews — Scholar-powered news

Marios Georgakis

@mariosgeorgakis.bsky.social

630 followers 300 following 350 posts

Physician-scientist leading a lab @lmumuenchen.bsky.social, visiting scientist @broadinstitute.org | Writing about genetics, omics, deep phenotyping, precision medicine https://www.deepvasc.com/

www.deepvasc.com

Posts Media Videos Starter Packs

Pinned

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

I'm often asked how human genetic data can be used to validate drug targets.

In our new ‪@natcardiovascres.nature.com‬ paper, we provide an end-to-end framework for genetically validating IL-6 inhibition for atherosclerotic cardiovascular disease outcomes 🧵

1 7 12

Marios Georgakis @mariosgeorgakis.bsky.social · 2d

🔗link to the paper 🔓: www.deepvasc.com/s/georgakis-...

www.deepvasc.com

Marios Georgakis @mariosgeorgakis.bsky.social · 2d

Stroke prevention is more nuanced than that of atherosclerosis in other vascular beds due to heterogeneity in underlying etiology.

In this paper we discuss the implications of recent trials for stroke prevention, as well as future perspectives for under development anti-inflammatory treatments

1 1

Marios Georgakis @mariosgeorgakis.bsky.social · 2d

If you're up for a niche read on challenges in translating anti-inflammatory therapies for atherosclerotic stroke prevention, have a look at our review in Neurology

1 1

Marios Georgakis @mariosgeorgakis.bsky.social · 26d

A great read👇

Some social science genetics papers are among the most interesting and methodologically rigorous I've read.

From biology to mating choices and inequalities, they deal with very fundamental concepts of what makes as humans.

Abdel Abdellaoui @dr-appie.bsky.social · Sep 8

Some people call it a minefield. Others call it dangerous, even irresponsible. I call it the most promising field in life sciences.

My love letter to social science genetics: communities.springernature.com/posts/a-love...

A Love Letter to Social Science Genetics

Some people call social science genetics a minefield. Others call it dangerous, even irresponsible. I call it the most promising field in life sciences.

communities.springernature.com

2 1 6

Reposted by Marios Georgakis

Kaur Alasoo @kauralasoo.bsky.social · Aug 27

Extremely proud to see this work led by @mihkeljesse.bsky.social now out on biorxiv!

mihkeljesse.bsky.social @mihkeljesse.bsky.social · Aug 27

After 1.5 years of work in @kauralasoo.bsky.social’s lab, we finally published my preprint! We introduce gpu-coloc, a GPU-accelerated implementation of coloc, show comparability to CLPP and aim to provide practical guidelines. Now accessible on BioRxiv: www.biorxiv.org/content/10.1...

Ultra-fast genetic colocalisation across millions of traits

Colocalisation is a powerful approach to assess if two genetic association signals are likely to share a causal variant. However, association analyses in large biobanks and molecular quantitative trai...

www.biorxiv.org

1 4 14

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Great work by Lanyue Zhang, Murad Omarov, and Lingling Xu.

Grateful for the partnership with @tourmalinebio.bsky.social (developing an anti-IL6 antibody) and the collaboration with
Pradeep Natarajan.

🔗link to study (open access🔓): www.nature.com/articles/s44...

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

However, each target and indication have unique nuances that necessitate individual tailored approaches for proper validation.

A scalable, one-size-fits-all approach risks overlooking important context and already acquired knowledge.

1 1

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

3⃣ Large-scale retrospective analyses of target-indication pairs (e.g. Minikel et al, Nature 2024) offer big picture evidence that genetically supported targets are more likely to lead to approved drugs.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

2⃣ Nearly all data used in this study are publicly available.

In principle, this framework could be applied to genetically validate any drug target, particularly those developed for chronic, aging-related diseases.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

We hope this work will provide prospective validation for the use of human genetics to predict trial-relevant clinical outcomes.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Thoughts for broader implications:

1⃣ Often, human genetic studies, replicate trial findings retrospectively.

The results of ZEUS, a phase 3 cardiovascular outcomes trial testing ziltivekimab in patients with ASCVD and high CRP, are expected in 2026.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Regarding side-effects, notable and consistent signals included:

👉 Glaucoma
👉 Perinatal & postpartum maternal hemorrhage

These may be relevant for specific populations and warrant further investigation.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

The main risk-lowering effects of the IL-6 proxy were replicated across:
👉 cardiovascular & metabolic endpoints
👉 autoimmune disease outcomes
👉 respiratory infections (pneumonia, influenza, COPD-related)

We also observed novel protective associations with:
👉 depression
👉 gallstone disease

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Finally, we performed a phenome-wide association study (PheWAS) in FinnGen to systematically assess whether our IL-6 genetic proxy is associated with any additional outcomes beyond those already studied.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Interestingly, our IL-6 genetic proxy showed no significant increase in the infection endpoints tested.

On the contrary, there was even evidence of risk-lowering effects on pneumonia hospitalization.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

The primary safety concern for anti-IL-6 therapies is infection risk.

In previous work, we have shown that genetically proxied IL6R inhibition increases risk of bacterial infections, consistent with clinical experience with tocilizumab.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

We also observed metabolic effects:

👉an association w/ lower risk of type 2 diabetes
👉significant increases in HDL particles

Across multiple metabolomic measurements, the effects were highly consistent with those of an IL6R instrument, indicating convergence downstream of IL-6 signaling

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

These findings were highly consistent in East Asian individuals from Biobank Japan.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

We then tested associations with cardiovascular endpoints.

Like an IL6R instrument, IL6 perturbation was associated with lower risk of atherosclerotic outcomes:

👉 Coronary artery disease (CAD)
👉 Peripheral artery disease (PAD)
👉 Large artery atherosclerotic stroke
👉 Carotid atherosclerotic plaque

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

In other words, our proxy predicts effects on biomarkers measured in the trial, with a correlation of 0.9 bet/ genetic & trial effects!

Similarly, the IL-6 genetic proxy showed effects on rheumatoid arthritis & polymyalgia rheumatica, for which IL-6 inhibition has been proven clinically effective.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

When indexed to CRP reduction, the effects of our genetic instrument on 8 tested biomarkers (data leveraged through publicly available GWASs) were highly concordant with ziltivekimab treatment!

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Can such a genetic proxy be used to study the effects of pharmacological IL-6 inhibition?

While phase 3 data are not yet available, phase 2 results from the RESCUE trial (ziltivekimab) provide biomarker evidence.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Indeed, most variants showed proportionally concordant effects on circulating IL-6 protein levels, supporting an effect on IL-6 abundance.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

All selected variants were non-coding, yet all were associated with IL6 expression in one or more cell types or tissues. This suggests they may modulate downstream signaling by altering IL6 RNA expression.

Marios Georgakis @mariosgeorgakis.bsky.social · Aug 27

Aggregating these 12 variants into a genetic score was associated with lifelong reductions in CRP (up to 24% for the upper vs lower percentie), comparable in magnitude to those observed with an IL6R genetic proxy.