Efficiently obtaining genomic transcript levels of thousands of patient samples by cancer types

Question

[theodore.killian]

I would like to observe the average transcript levels of specific genes within specific types of cancer. Obviously, downloading individual BAM files and sequencing them is not feasible. Even downloading count matrix files and normalizing them would be a big hassle. I'm looking for a database where I can easily download the thousands of genomic expression levels by cancer type. Ideally, average genomic expression levels by cancer type would be even better, but individual patient data will do.

Can anyone point me to a database that would be able to provide this kind of data?

For example, I want to reproduce something like Fig 1D in this paper, but for different genes: https://www.sciencedirect.com/science/article/pii/S2211124720307877