Early cancer detection tools mostly involve circulating DNA (or cell-free DNA (cfDNA)) in the bloodstream. For instance, companies, such as Grail and Thrive Early Detection, have developed tools that look at the methylation marks and mutations of cfDNA, respectively, to detect multiple cancers early. A blood test is relatively noninvasive and early diagnosis is key to successful cancer treatment in general. Thus, these blood tests (also called liquid biopsy) gained attention in the past few years.
If you think liquid biopsy is all about biomarker molecules floating in the blood, such as cfDNA and protein, you are missing something. A recent paper by Sol et al. published in Cell Reports Medicine demonstrates that RNA profiles of platelets in the blood could be used to detect a brain tumor, glioblastoma.
Platelets, the smallest of blood cells and the second-most abundant cell type in peripheral blood, are known for their roles in forming blood clot to stop bleeding. Sol et al. showed that platelets collected from glioblastoma patients have unique spliced RNA profiles, which could be a source for diagnostic.
The fact that platelets respond to the tumor environment is fascinating and demands more research on the underlying mechanisms. With a VC hat on, I was wondering how much blood was taken from each patient to get meaningful data. 2mL? 10mL? Another question is: can this RNA sequencing on platelets be done in conjunction with other blood tests that are rather interested in blood plasma? Just like Grail's and Thrive Early Detection's tests, Sol et al.'s study used sequencing. What types of new (and possibly faster and more accurate) sequencing techniques are being developed? I guess that would be my next project.
I don't worry too much about false positives because scientists are fully aware of them and improving ways (e.g., combining sequencing with MRI) to minimize false detection. One thing we cannot avoid talking about when we talk about cancer diagnostics is screening guidelines. Jocelyn Kaiser wrote in Science, "Will decade-plus studies that show survival benefits be required, for example, as they were for other screening methods such as mammograms and lung scans for smokers? Or can proxies, such as data showing the tests find more early cancers than existing screening methods, satisfy the various groups?" It reminds me that following up on the regulation side of biotech is as important as evaluating the technology.