Comment by maxander | Hacker News Reader

Comment by maxander | Hacker News Reader

Parent Full thread

An old XKCD is, as usual, highly relevant: https://xkcd.com/1217/

Granted, this is actually a trial in humans, so that's good! But if you look at the study, the focus is on detecting circulating cancer cells; attempting to use the technology to destroy them is relegated to one paragraph near the end. There they note that this seemed to happen by accident, to a limited degree, in 6 out of the 18 patients; so they tried again in a single patient, whose melanoma cell melanin content was particularly favorable for this process, and it again seemed to work ("up to a 48-fold reduction" in circulating melanoma cells.)

They went on to do in vitro tests to try and show it would work in general, aiming the laser at a test tube full of blood; this worked, which is encouraging, but it shouldn't be taken as a strong indicator of anything practical. (And even there, they note that the melanin content of the tumor cells is important to the treatment's effectiveness.)

And yes, it's only targeting circulating melanoma cells. This tech can't just aim a laser at a tumor and kill it; it probably isn't useful for non-solid cancers like leukemia, either. So while this is cool and interesting, it's not a huge breakthrough in cancer treatment.

The XKCD is spot-on here.

> “In one patient, we destroyed 96 percent of the tumor cells” that crossed the laser beam, says Zharov.

Just want to clarify: tumors will generally drop hundreds of thousands to millions of cells into the blood stream on a daily basis. Most of the circulating tumor cells are already unlikely to become mets at baseline, and by the time you've found a tumor you've already seeded these cells in a number of locations throughout the body. This seeding usually occurs long before we can detect the primary tumor. We get "metastasis" when a number of other factors come into play that allow a particular CTC / bundle of CTCs to start growing into a met. For those interested, R. Weinberg's text on cancer is pretty much the authority, and he discusses this at length.

As it exists, as a therapy, it's for catching some of the horses that are making it out of the barn - after the doors have been open for a while.

To be useful as a therapy, we'd need something that can be worn 24/7 - if I catch a tumor that needs to be excised, I'd put it on your arm, and use it to deplete the circulating population to reduce the likelihood of metastasis happening in the period between detection and excision. And I'd need an RCT to show me that this is actually going to result in a reduction of metastasis, since there's every likelihood that the eventual met won't come from the CTCs seeded in the period between detection and excision.

Don't get me wrong; I'm happy to see more ways to detect and kill tumor cells. But this is still many steps away from becoming a technology with clinical application. At this stage, its relationship to useful cancer therapy is the relationship between my laser pointer from ten years ago and a sci-fi laser pistol.

What this is cool as, is as a mechanism for helping detect cancers. I'm all for having something I can slap on your arm and let me know if you've got melanoma brewing, rather than having a dermatologist manually inspect all your crevices for an overlooked mole.

Reply on news.ycombinator.com