Harvard Scientists Create Hand-Held Device to Detect Cancer at Bedside

I'm a pathologist.

  "This is superior to even the pathologists," Castro said.

This is a stupid statement and a ridiculous conclusion in the linked article. Let's talk about why this is a shitty conclusion.

1. This study was looking at protein markers on cells that were aspirated out of stomach masses and comparing the results to the pathology report. This thinking completely discounts the other actor in the scenario, the gastroenterologist who is doing the aspiration by endoscopy. If you give me a sample with cancer, I'm going to report it. However, the GI doc has to sample it and he has to give me a decent sample to work with. If he does not biopsy your cancer, I can't call it. I don't practice faith-based medicine. The thing is though, I spend a not-insignificant portion of my days in procedures just like this immediately assessing the material (looking at it under the microscope) and can tell the GI if he is sampling the material, or if it is enough for us to make the diagnosis. He usually knows if it is cancer before I leave the room (and it is a lot shorter than even 60 minutes).

2. This small device has detected a single type of cancer in pre-screened patients, that is patients who have a reason to have this done (maybe they're just having a little stomach pain, or maybe they've been vomiting blood and have had a 20 pound weight loss in the past 3 months). Try screening the general population and let me know how your device fares. It is not portable. It required an entire endoscopy suite, expensive endoscopy equipment, and a highly trained (multiple years of clinical fellowship) operator. You ain't gonna pull this off in a rural village in Africa (and let's face, what's the point of that, you don't have anything to treat it with). Plus you were directly sampling the tumor mass with a needle. I do this all damn day! I don't really have much problem calling cancer, or even better, telling docs that masses are benign if someone is able to stick a needle in something. Wake me up when you can diagnose any form of cancer by sampling tumor cells circulating in the peripheral blood (actually we are probably closer to that than you think!).

3. For lesions of the stomach, aspirating the lesion is a shitty way to make the diagnosis. Forceps biopsy is a much better way and that is the way this type of cancer is going to be diagnosed in the vast majority of patients. So if you are already there with your endoscopy, just get a goddamn biopsy already.

This is a great finding. I'm doing braincancer research myself and that's where my main concern comes from. By reading the abstract (on a train, no time for the full article yet) I understand they are using fineneedle aspiration to get their samples. That would work for most easy accessible tumors but for braintumor the human skull will be too big of a barrier. I'm still hoping for methods where samples from the peripheral blood will be used.

Secondly I wonder what the costs willbe when antibodies for more than just one specific tumor will be used. Now they scan for one tumor by using 4 proteins. Scanning for more types of cancer will increase costs dramatically thought.

Nevertheless great finding and hopefully available in clinics all over the world soon (but it usually takes an other 10 years).

The most fascinating portion of it, to me:

"The technology is also more accurate than current diagnostic tests, the scientists said. The portable diagnostic device achieved 96 percent accuracy, compared with the 84 percent rate associated with more traditional assays."

Not only is it significantly smaller and much faster, it is also more accurate. Let's hope we see this deployed to the locations where it is most needed.

The article mentions the cost of the device itself, but I wonder how much it costs to administer the test. Are there any consumables that must be used? How much do they cost?

I heard a talk about this technology about 4 years ago from MRI/NMR researchers at my university working on it. The big problem is getting the right biomolecules that will bind with the target cancer. This is apparently a common problem in many scenarios, especially targeted therapies. Other than that the technology sounds very interesting, as you can potentially detect a single cancer cell in the sample.

Here is the link to the paper http://stm.sciencemag.org/content/3/71/71ra16.full.pdf

With relay back to a previous thread on 'advances seen in our lifetime' it again amazes me how fast things are moving, especially in the medical field. And for $200 yet... brilliant. Leonard "Bones" McCoy would be proud.

just wanted to share with anyone interested in cancer.. I finished reading this amazing book "The Emperor of All Maladies: A Biography of Cancer" Great for anyone but specially those outside the field who want to know abt the history of cancer.

http://www.amazon.com/Emperor-All-Maladies-Biography-Cancer/...

The one bit of information left out in the article is how do you get the magnetic nanoparticles in your blood stream in the first place. I wouldn't think they are injected directly into a vein, but maybe they are ...

It is one of my dreams to be involved with cancer, perhaps a medical device startup. One of my college professors (biology of cancer) urged me to go to med school but cancer specifically has always been very attractive to me. If I wanted to get involved with a biotech/med device startup where would I go about looking?