Making A Breakthrough Better: Improving Low Dose CT Scans

Making A Breakthrough Better: Improving Low Dose CT Scans

CREATED May 6, 2013

DURHAM, N.C. (Ivanhoe Newswire) - It's a technology that's helping dramatically reduce lung cancer deaths. Low dose CT scans have revolutionized early diagnosis and early treatment, but sometimes it's hard to tell if the abnormalities they detect are really cancer. Now, researchers have discovered something in our blood that could make the scanner even better.

An internet search may have saved Lawrence Moore's life.

"I just happened to run across the article," Lawrence Moore told Ivanhoe.

The article about low dose CT scans detecting lung cancer motivated the former smoker to get one done. A mass was found, but it was inconclusive. Lawrence had to wait seven months and go through more scans before he was finally diagnosed with lung cancer.

"You're sitting there sort of gnawing your fingernails and wondering what's happening," Lawrence said.

Duke University's Dr. Edward Patz said it happens to many patients and can mean more testing or invasive biopsies, but his team discovered three blood proteins that were found to be higher in patients with cancer.

The doctor said when an abnormality is inconclusive images from a low dose CT scan that show its size and the blood biomarkers, can help determine if lung cancer risk is high or low. The test can be completed in about a day.

"It could help patients avoid some unnecessary procedures and not delay treatment in other patients," Edward F. Patz, Jr. MD, James and Alice Chen Professor of Radiology at Duke University School of Medicine, told Ivanhoe.

Lawrence had a portion of his lung removed within 24 hours of his diagnosis.

"I was very glad it was over with," Lawrence said.

He's cancer free and his biggest worry these days is watering his orchids.

The doctor said using the blood bio-markers and images together is 80 to 85 percent accurate in determining if an inconclusive abnormality is cancerous. It is FDA approved, will cost around $100, and is just about ready to be rolled out. Dr. Patz said the labs that will handle the test are currently getting their infrastructure ready for it.


BACKGROUND:   Lung cancer is the second most common cancer in both men and women, accounting for about 14 percent of all new cancers.  The American Cancer Society estimates that in the United States in 2013 there will be over 200,000 new cases of lung cancer.  There will also be an estimated 159,480 deaths from lung cancer, about 27 percent.  Lung cancer is the leading cause of cancer death among men and women.  It mainly occurs in older people.  About two out of three people diagnosed with lung cancer are 65 years or older.  The average age at the time of diagnosis is about 71.  The chance of a man developing lung cancer is one in 13; for a woman the risk is about one in 16.  (Source: www.cancer.org)

RISKS:  Patients at high risk for developing lung cancer are those who are between the ages 55 and 74, are current smokers or quit within the last 15 years, or smoked for 30 or more pack years (pack years means the average number of packs smoked per day multiplied by the number of years a person has smoked.  (Source: www.seatlecca.org)

LOW DOSE CT SCAN:  Early detection is proven to be a successful strategy for fighting many forms of cancer.  More than 75 percent of people with lung cancer have incurable, locally advanced, or metastatic disease at the time of diagnosis, with a five-year survival rate less than 5 percent.  Detecting lung cancer at its earliest stage and having it removed surgically can mean that a person can expect a five-year survival rate that is closer to 70 percent.  During a randomized national trial involving more than 53,000 participants who were current or former heavy smokers between the ages of 55 and 74, the low-dose helical computed tomography (CT) found 20 percent fewer lung cancer deaths among trial participants screened with low-dose helical CT. Helical CT uses X-rays to obtain a multiple-image scan of the entire chest during a 7 to 15 second breath-hold.  A standard chest X-ray requires only a sub-second breath-hold but produces a single image of the whole chest in which anatomic structures overlie one another.  It was the first time that clear evidence of a significant reduction in lung cancer mortality with a screening test in a randomized controlled trial was discovered.  (Source:http://www.cancer.gov/newscenter/newsfromnci/2010/NLSTresultsRelease)

NEW TECHNOLOGY:  Researchers at Duke developed a new diagnostic test that employs an instrument called "MALDI-TOF MS" to detect proteins in the blood that signal inflammatory diseases and cancers.  By finding a disease-causing protein, doctors can diagnose the disease and develop new ways to block its detrimental effects.  The specific protein, serum amyloid A, was elevated in the blood of lung cancer patients but not in the blood of normal patients.  Serum amyloid A has previously been shown to be elevated in cancers and other diseases, but researchers at Duke were the first to use MALDI-TOF MS to identify this protein and others that may be involved in lung cancer.  Once the proteins are identified, they can be used as biologic markers to diagnose the earliest stage of cancer, possibly before a CT scan can pick up the image of a tumor on the lungs. Based on their findings, researchers plan to develop a blood test that will measure serum amyloid A and other specific proteins that can detect lung cancer in the blood before a tumor is apparent.  (Source: http://www.dukehealth.org/health_library/news/7037


Edward Patz, MD, James and Alice Chen Professor of Radiology, Professor in Pharmacology and Cancer Biology at Duke University School of Medicine, talks about how blood biomarkers and imaging technology can be used together to diagnose lung cancer.

How important have low dose CT scans become in helping diagnose lung cancer? 

Dr. Patz: CT scans are used on an everyday basis and they are among the tools that we use to try to detect lung cancers early. There is the recent National and Cancer Screening Trial, which did show a reduction in lung cancer mortality, but now we are trying to figure out how to efficiently put that into large public health screening programs.

What have you discovered that has made CT scans even more effective? 

Dr. Patz: One of the biggest problems with the low dose CT screening trials was that a large percentage of individuals enrolled in those trials had indeterminate pulmonary abnormalities, meaning they had an abnormality but we were not sure if it was cancer or not cancer. That occurred in almost 25% of participants. So, we see a large number of individuals who have these abnormalities, but only a small percent will actually have lung cancer.

What happens when you do see something that is abnormal, but you cannot determine at that point if it is cancer or not? Does that person have to go for more testing and maybe invasive surgery?

Dr. Patz: Almost all of those patients will have further evaluation. They could have sequential studies or multiple sequential studies at 3- to 6-month intervals. They may go for an additional test, such as a PET scan, which helps us differentiate a benign tumor from a malignant abnormality. Some of them may go for an invasive procedure, such as a biopsy, or may even go directly to surgery.

How does your blood biomarker work? What does it do? How does it supplement the CT scan so people can maybe avoid those procedures?

Dr. Patz: This test draws the patient's blood. We look at the radiograph, either a CT or a chest film, because we need to know the size of the nodule, which tells us something about the likelihood or the risk for lung cancer. We use both to come up with the patient's risk of lung cancer, whether it's high, moderate, or at low risk for lung cancer. At this point we have the abnormality, we draw the blood test. Then we can develop risk stratification so we can more efficiently guide patient management.

What is in the blood that you look for?

Dr. Patz: We are looking for three separate proteins and these are proteins we discovered in the lab, which we found to be higher in patients with cancer rather than those patients without cancer.

Was that something you guys discovered here?

Dr. Patz: Right, we discovered it right here in the lab. 

Would you take the CT image or a chest x-ray? 

Dr. Patz: It could be a chest x-ray or a CT image. We combine the two with the blood biomarkers and we come up with a risk stratification.  

How does the procedure work once you have all the information? You have the CT scan and the blood work, what happens next? How long will it take to get the information back whether it is high or low risk?

Dr. Patz: Oh, that is very fast. I mean it is back within a day or so. It is simple. We already have the assay worked out and then we provide the clinician, whoever ordered the test and is taking care of the patient, with that information. They can then decide on the best way to manage the patient from there.

Is this widespread at this point? Are you hoping to be widespread? 

Dr. Patz: No, we have not really introduced it. The Phase 3 biomarker studies have been completed and published. So, now we are just trying to introduce this into clinical practice. We will see what the adoption rate becomes. 

So you just kind of got this new toy and you just hope more people will use it?

Dr. Patz: Correct. We think it will be quite useful to manage these patients, which right now everybody's fundamentally managed the same. There really is not a good way to separate them out and then some patients may avoid the risk of biopsy. Instead of 3 months, they may delay a CT scan to a year and we can then reassure them that the risk is fairly low. The test is not perfect. We have to tell patients that it is not absolute, but at least it gives us better guidelines than we had before.

How accurate did it turn out to be in determining the low- and high-risk patients?

Dr. Patz: The overall accuracy of the test is around 80% to 85% accurate. So for the vast majority of patients we are correct, but that still means approximately 15% to 20% of the patients will not be accurately determined. We believe that if they have cancer, the risk is still very low that they will progress in the interim because we are very conservative about the recommendations. So that they will not progress and it really should not have any detrimental effect to their long-term outcome.

In that 20 percent to 15 percent, did it skew toward diagnosing cancer or did it skew toward not catching the cancer? 

Dr. Patz: We tried to catch as many lung cancer patients as we could with the understanding that we have more so-called false positive tests. The vast majority of cancers that we did not pick up were very small nodules, but they were all in the early stage disease and all treated exactly the same.

If this is adopted, what is the procedure? Do they have to send it to Duke specifically or do can they do it from their own office?

Dr. Patz: No, this will be a laboratory-based test where they will be able to send it into a central laboratory and they will perform the test.

So any laboratory they use for other blood work, they can use for this?

Dr. Patz: There probably will be specific laboratories. At this time, the study was performed with one of the central labs, which was LabCorp. They are the ones who are developing the test, but they have offices all over the country and they can perform the test anyplace.

Without this, if you saw an abnormality, would it take a lot longer than a day to help a patient figure out if they had a high or low risk?

Dr. Patz: There is no other way that we know of to determine whether they have a high or low risk. Right now, we use an algorithm based on size as to how we should manage them. However, this actually gives us additional confidence to see if what we are doing is correct and it could actually reduce the risk of scans. It may give the patients more confidence that they do not have lung cancer or it may tell us that we need to act sooner rather than 3 months and we may need to be more aggressive about an abnormality. 

So it could help speed up treatment if need it or help people avoid it if they do not need it?

Dr. Patz: That is right.  

How long did it take to develop this?

Dr. Patz: It has taken us several years to do the discovery and the initial validation. Then we just completed the Phase 3, randomized biomarker trial, or the blinded independent study on a prospective group of patients, and that took several years. So, overall, it has probably been about 8 years for us to develop the test.

Did the biomarker help Lawrence get treatment faster?

Dr. Patz: I do not know. I cannot tell you if he was. I know he was one of our patients who consented and we have his blood, whether he was in that prospective trial, I do not know right now. No he was not. I can tell you, that he did not help him because at the time of the trial you really cannot use the biomarker to change management until you know that it is really effective.

He is going to be an example of what it is like before this?

Dr. Patz: Yes. He is a patient who would have benefitted from this type of marker.

So, is it being used anywhere at this point?

Dr. Patz: No, it is not. We have not rolled it out yet, but we are in the process of doing that, trying to make sure that the infrastructure is there.

Do you have any idea when it might be rolled out?

Dr. Patz: That is more of LabCorp's decision rather than my decision at this point. We performed the trial. We developed the markers and performed the trials. Now, it is up to them to help roll this out and make available for clinics. 

What is that infrastructure that needs to be built? Can you kind of describe what needs to be in place in order for this to be?

Dr. Patz: Well, first of all they need to make sure that all their laboratories can perform the tests. Several of them can now, but need to make sure when they collect the blood that they know that it is part of this lung cancer management protocol.  Hopefully this will make it to where at the end they give the results of the specific markers, meaning the blood level, but they can also use this algorithm that are developed by combining the nodule size with the markers to come out with a risk stratification. That sort of package has not been fully commercialized at this point.

What are you looking for specifically in the proteins? Is it an elevated level of the proteins?

Dr. Patz: It is an elevated level of the proteins. We found 3 proteins in particular, which were elevated in patients with lung cancer as compared to patients who did not have lung cancer.


Edward Patz, Jr., MD
James and Alice Chen Professor of Radiology
Professor in Pharmacology and Cancer Biology
Duke University School of Medicine
(919) 684-7367