Blood tests for cancer may be possible in the near future

Most cancers that are solid tumors (like lung, breast, thyroid, bowel and others) can’t be detected by blood tests. Until now.

A few tumor types do release substances into the bloodstream that are detectable. For example, some tumors of bowel (and a few other sites) release carcinoembryonic antigen (CEA). So that might serve as a method of detecting those types of tumors. However, CEA can also be elevated in non-malignant diseases (some cases of cirrhosis, pancreatitis, ulcerative colitis and others), and is also elevated in cigarette smokers. So while finding an elevated CEA level is a clue, additional testing, like a CT scan, is often necessary.

Detecting a cancer often comes down to the patient’s clinical history, family history, social history, the doctor’s physical examination of the patient, and often some form of imaging tests like an X-ray, ultrasound, CT or MRI.

Woman with crossed hands, via Shutterstock

Woman with crossed hands, via Shutterstock

In the recent past, researchers have tried to find tumor DNA circulating in the bloodstream. DNA from malignancies is thought to enter the bloodstream when tumor cells die and break up. Even tumors that haven’t been treated with chemotherapeutic drugs or radiation still have cells dying. The dead cells’ DNA joins DNA from other, non-cancerous cells that is also circulating. So finding tumor DNA in this mass of non-tumor DNA is daunting. Researchers looked for bits of tumor DNA that would have come from a single mutated gene that was known to be associated with a particular type of cancer. As an analogy, try looking for a needle in a haystack after another, larger haystack has been dumped on top.

The scientists in this paper made use of a somewhat different technique. They focused on non-small cell lung cancer (NSCLC), a common type of cancer. Right now, there is no marker in the blood that will allow NSCLC to be detected. Instead of looking for bits of DNA from just a single mutation, these researchers looked at the bigger picture. They identified 139 mutations, some of which can be found in most non-small cell lung cancers. Then they looked to see if any fragments of DNA from any of those mutations were present. They found that they were able to detect cancers in all patients that they tested who had stage 2 or higher cancers. And, they were able to detect cancers in 50% of patients who had stage 1 cancers.

In the original research, they were only looking for DNA from a single mutant gene. Not all of the tumors would have that particular mutation. So the numbers of tumors that they could detect would be lower, i.e., they could only detect NSCLCs that had that one mutation. They would miss all other NSCLCs that had any other mutations, but were lacking in that one single mutation that they were set up to look for.

In the cited research, they looked for DNA from 139 different possible mutations, any of which could be found in NSCLCs. So they could detect a much larger number of tumors, since they were looking for a broad spectrum of different mutations that were found in those tumor types.

In both cases you’re looking for DNA that has a sequence of amino acids that is specific for a given mutation (in this case one of the 139 that they were looking for). In the original research, you’re only looking for one specific DNA sequence. In the cited assay, you’re looking for any one of those 139 mutations.

The test has good sensitivity and specificity. Meaning that there are very few false positives and false negatives. So that if any tumor DNA is detected, it’s highly likely that there really is NSCLC present. Conversely, if none is detected, it’s highly unlikely that any NSCLC is there.

Cure rates are usually higher the lower the numeric stage of the cancer. So, this method could be used as a test to see if someone has NSCLC. Being able to have a blood test that serves to detect cancer could be a wonderful breakthrough.

Their research also showed that their technique may allow them to detect tumor recurrence. After treatment, the amount of tumor DNA circulating decreases. If the tumor returns or begins to grow, the amount of tumor DNA in the blood increases. Currently, without a good blood marker for the tumor, patients often need to undergo repeated CT, PET-CT or MRI scans to see if there is evidence of tumor or tumor spread. Those are costly tests. Following the patient with a blood test would be simpler, more convenient and less expensive.

They’ve also found some other possible applications for their technique, as well. One might allow them to tell if a specific chemotherapy drug might not work against a particular tumor type. This could save the patient from undergoing one type of chemotherapy that might not work against the particular tumor that they have.

And the investigators see no reason why similar techniques might not be used against many other types of solid tumors. Some tumor types, like ovarian and pancreatic cancers, are notoriously difficult to detect with current methods. This test might allow for earlier detection of these tumors, and others, as well. If it becomes possible to test for other cancers using this method, perhaps a single blood test could be used as a routine screening tool for several cancer types. For example, a single blood specimen might suffice to test for cancers of the lung, breast, ovary, bowel, pancreas and other tissues. But all of this is in the future. Some of it might be in the not-too-distant future.

Currently, these researchers are working on clinical trials for further evaluation of the technique in testing for NSCLCs. They are also looking into applying the technique to other types of solid tumors.


(NOTE FROM JOHN: It’s hugely important to our continued success that you share our stories on social media by using the “Like” buttons at the top and bottom of this story to share it on Facebook, Tweeting it to your friends, and sharing it on other services. Without that additional traffic, our advertising dies, and so do we. We need your help – if you like one of our stories, please share it online.)


Mark Thoma, MD, is a physician who did his residency in internal medicine. Mark has a long history of social activism, and was an early technogeek, and science junkie, after evolving through his nerd phase. Favorite quote: “The most exciting phrase to hear in science... is not 'Eureka!' (I found it!) but 'That's funny.'” - Isaac Asimov

Share This Post

  • quax

    Excellent news, and very informative as always. These posts have quickly become my Amaricablog favorites.

  • Barry at Genesys Biolabs

    Dr. Thoma, Are you aware of a new blood test called Paula’s Test that looks at a panel of tumor antigens in the blood and can identify lung cancer at the early stage with 74% sensitivity? It is now on the market from Genesys Biolabs.

  • http://heimaey.us/ heimaey

    Thanks. It was almost 9 years ago now, but I remember them ruling out cancer because some indicators weren’t there…then…

  • docsterx

    Oh, heimaey, that’s terrible. I’m sorry.

  • http://heimaey.us/ heimaey

    They have some kinds of tests now but they’re shitty. My boyfriend tested negative for cancer and died two months later from it.

  • docsterx

    The testing is still only being done as research, not available in commercial labs, as yet. So, it’s not available to the general public. Since it’s research, those participants in the studies are tested free of charge. What the cost will be when it’s available, is anyone’s guess. One of the researchers said in an interview that they were trying to think of ways to keep costs down, so that’s promising.

    How long it would testing take? Only a guess, but probably several days after getting the blood specimen at least at this stage in the research.

    Thanks. But I was kind of getting attached to ZombieDoc.

  • Cletus

    How expensive are and how long do DNA tests take these days, Dr Mark? (P.S. – thanks for taking the avatar advice ;-) )

© 2014 AMERICAblog News. All rights reserved. · Entries RSS