Thread by @simocristea

Today's amazing science dives deep into the 2 strongest #cancer modulators: evolution & immune defense.

First-ever detailed temporal evolutionary trajectories for 600,000 B cell lymphoma immune cells #scRNAseq & #scTCRSeq of 32 patients during immunotherapy with 2 CAR-T drugs 🧵

First, what is chimeric antigen receptor (CAR) T cell therapy?

It is an immunotherapy in which the patient's own immune cells are genetically engineered ex-vivo to recognize, attack & kill tumor cells. Then they are infused back into the patient, ready to fight the enemy!🤺 2/13

Immunotherapies have revolutionized cancer treatment & are among the most promising future approaches.

However: response rates, even if varying across cancers, remain limited, with e.g. 50% response in lymphomas.

Why such therapies fail for the other half remains a mystery 3/13

This study analyzed 32 large B cell lymphoma patients treated with 2 CAR-T drugs: axicabtagene ciloleucel (axi-cel) & tisagenlecleucel (tisa-cel). During the 6 months of the study, each patient was sampled before, during & after therapy. At the end, half of patients relapsed 4/13

By analyzing the differences between how the CAR-T infused product (IP) interacted with the patient's own immune cells 7 days post therapy (D7) in patients who responded vs. who didn't, one can chart mechanisms by which the immune system succeeds or fails in curing cancers. 5/13

The main finding is that these 2 commonly used drugs have strikingly different effects on the immune system: tisa-cel responses were associated with huge expansion of rare CD8+ central-memory-like cells that were infused, whereas axi-cel treatment showed less Tcell shifting 6/13

Even more: axi-cell nonresponders had strong enrichment in CAR-T regulatory cells (T regulatory cells with a CAR transcript attached).

Since baseline levels of Tregs were similar for patients in both treatments, this enrichment was not due to pre-existing differences. 7/13

Why is this?Both drugs are commonly used & similar in purpose.

However: they have different designs & manufacturing processes & are delivered by different vectors.

The authors hypothesize that starting manufacturing from fresh (axi) vs frozen (tisa) cells might play a role 8/13

Therefore: cryopreservation might imply strong depletion of Tregs, which are intolerant of freezing.

This hypothesis is interesting & has far-reaching potential implications for understanding the success/failure of different immunotherapies across other cancer types as well 9/13

Next, in a beautiful controlled in-vivo experiments, the authors prove‼️ the causality of how these newly evolved T regulatory cells suppress the expansion of the “good” infused T cells & facilitate CART relapse.

This is a first in showing that Tregs influence CART relapse 10/13

How is this done?

Mice were engrafted with lymphoma cells followed by injection of CAR-T cells 7 days later, made of 95% CAR-Tconv with either 5% CAR-CD4,or 5% CAR-Tregs.

All tumors were cleared by day3

But then: all CAR-Treg mice relapsed,while none of the control ones did🤯

How impressive 😲

CAR-Treg cells making up only 5% of total infused CAR-T cells were sufficient to drive tumor relapses and suppress CAR-Tconv expansion.

Such science is amazing because it shows mechanistic routes of relapse in one of the most promising cancer therapies. 12/13

Such data can further be mined to gain hidden insights via e.g. ML models comparing responders vs. nonresponders.

To download data:
ncbi.nlm.nih.gov/geo/query/acc.…

Jupyer notebooks:
github.com/getzlab/Haradh…

Paper by the labs of M. Maus, Cathy Wu & @getz_lab 🙏:
nature.com/articles/s4159…