As a disclosure, I do a good number of biopsies though it is not material to my business and to be perfectly honest there are much more efficient uses of my time. Also I have only a superficial understanding of what GH does.
The article that Saul has posted (https://www.statnews.com/2019/03/18/liquid-biopsy-patient-pe…) regarding the patient’s experience with a lung biopsy is certainly real but far from typical. It is a combination of an overly sensitive/anxious patient possibly with a doctor with poor communication skills (or dramatized to promote liquid biopsy research) with a (presumably) technically difficult biopsy and a cancer that is known to be difficult to fully diagnose with needle biopsy.
- The description of the procedure (pushing the needle…near the heart…make sure it’s in the right place) it not how I would describe it to a patient because it sounds bad. We know where the important structures are, we are just making slight adjustments. With the lung, it moves around so there’s a bit more to it but still nobody is going to biopsy something that puts the heart at real risk.
- Lymphoma is notorious for being difficult to diagnose with biopsy, and a lot of doctors tend to use smaller gauge needles to do lung biopsies which just adds to the problem. As I understand it the architecture of the mass matters to pathologists.
- The risk of pneumothorax (air getting around the lung, or to put it in a good scary headline a “collapsed lung”) is often quoted at 20%. In my experience it’s much less unless you just have incredibly ill patients or poor technique. Sometimes you get a bit of air getting in around the lung because you’re entering from the outside, but those have no clinical significance. In the patient’s case she had to be admitted but I tend to see something like 10% pneumothorax rate with 80% of those patients going home later that day after a follow up chest x-ray shows it’s stable.
Anyway, my general point is that tissue biopsies except in certain locations are overall very safe in a healthy enough patient. And one could argue that someone not healthy enough to go through a tissue biopsy probably isn’t going to tolerate whatever treatment might be needed anyway.
Regarding the potential for liquid biopsy, I am not sure how much detail they can get. Some cancers are stratified based on specific DNA mutations while others are based on expression patterns. A cell can have a certain mutation but if it’s turned on or off it may not produce a certain protein needed for specific targeting therapies. Does this show up with GH’s technology? Other times it may be the way cells are arranged or what not that might not be obvious based on DNA evaluation.
Presumably there will be some kind of false negative rate. There must be some point at which tumor DNA does not yet reach the bloodstream but we can see it on imaging. And will there be a false positive rate? Will we be searching endlessly for an invisible tumor if so? If a patient tests positive for a liver cancer that we can’t see, but there’s a lung nodule, are we obligated to biopsy the lung nodule now trying to figure out if it’s spread from the liver or a separate cancer?
With the low/no risk nature of the test, other than potentially going down the rabbit hole of chasing a false positive or treating something that would have never mattered (finding a small lung cancer in a 90 year old with bad emphysema and heart failure is kind of useless) there is certainly a potential market for it. I don’t believe it will displace tissue biopsy, at least for many years, and might actually increase the number of biopsies as it is being developed as a definitive test rather than a screening test (more people willing to go through with a blood test only to find cancers they never would have known about). There are plenty of sales to be had, and in the medical field the science is only a part of the equation as to whether a product is successful.
Since I can’t risk a meaningful amount of my portfolio due to the risks, it’s hard for me to invest (a 10 bagger on a 1-2% position is only as good as a double in a 5-10% position which I feel much more sure about). But I’m interested in it from a professional standpoint.
