Share on Pinterest

Monitoring changes in blood tests over time could better predict cancer risk than the results of a single blood test, a recent study has found. parichat wongyai/Getty Images
  • A recent study has found that monitoring changes in blood test results over time, rather than focusing on a single test, may help identify cancer risk earlier.
  • In people who developed cancer, changes in blood tests often occurred 3 to 5 years before they presented with unexplained weight loss.
  • Adjusting blood test trends by age and sex improved accuracy, with certain trends associated with specific cancers.

In primary care, blood test abnormalities can be one factor considered in cancer risk assessments. However, they generally primarily focus on whether individual test results are abnormal at a specific moment in time.

A study published in September in PLOS Medicine suggests that monitoring changes in blood test results over time may be more beneficial for assessing cancer risk than relying on information from the most recent blood test only.

The researchers focused on 275,205 patients presenting with unexplained weight loss, a possible indicator of cancer, among whom almost 14,000 received a cancer diagnosis within 6 months.

They aimed to determine whether trends in 26 common blood tests could help support an earlier cancer diagnosis.

“In some instances, monitoring how a patient’s blood test results change over time could offer further value,” said study author Pradeep Virdee in a press release.

The researchers found that changes over time in blood tests could help indicate cancer risk.

They investigated trends in the 1, 3, 6, and 10 years before the participants presented with unexplained weight loss. They found that, for those who subsequently developed cancer, changes in blood tests began to happen around 3 to 5 years before unexplained weight loss.

The study authors note that the changes were often still within the normal reference range. Exceptions to this included:

  • hemoglobin and red blood cells in patients ages 70 years and older
  • C-reactive protein (CRP)
  • erythrocyte sedimentation rate (ESR)
  • plasma viscosity
  • hematocrit and red cell distribution width (RDW) in adults ages 50 years and below

After adjusting blood test abnormalities and trends by sex and age, the area under the curves (AUCs) were higher compared with the unadjusted estimates on 34 occasions. The highest AUC was 0.82, a strong score where 1.0 is considered a perfect discrimination ability and 0.5 refers to no specific discrimination ability.

“We show how blood test results are made more accurate for cancer by adding patient age and sex. This relatively simple calculation could easily be performed at the laboratory,” said study author Brian Nicholson in the press release.

Ketan Thanki, MD, board certified colorectal surgeon who specializes in benign and malignant disease of the colon, rectum, and anus with the MemorialCare Todd Cancer Institute at Long Beach Medical Center in Long Beach, CA, spoke with Medical News Today about the study findings.

“First, interpret results in the context of age and sex, not against the lab’s reference range. This is the main finding. The same variation in an elderly patient may be less meaningful than in a young patient. Second, weight loss with an abnormal routine blood test is a real cancer signal. Five percent of these patients had cancer within 6 months! That’s a lot of people,” said Thanki, who was not involved in the study.

“Lastly, compare [the] patient’s values against their own over time, not necessarily against reference values. As a result, in designing programs to flag patients at risk of developing cancer, we should look at each patient’s own trajectory, not just the patient in a snapshot in time or as they relate to the reference values,” he added.

The study authors found that changes in some of these and other blood tests were linked with certain types of cancer.

For example:

  • white blood count and neutrophil trends were associated with lung cancer
  • mean cell volume trends were associated with bowel cancer and lymphoma
  • red blood cell count, aspartate, aminotransferase, hematocrit, and platelet-to-lymphocyte ratio trends were associated with prostate cancer

While these findings do not suggest that individual blood tests could diagnose a particular type of cancer, they could indicate that trends in blood test results over time could provide additional information about the risk of certain types of cancer.

The authors noted that the most commonly available blood tests were hemoglobin, white blood cell count, and platelets.

“These are not only commonly and easily available, but are very good markers for inflammatory states, blood loss, and nutritional deficits that may be linked to cancer,” Thanki explained.

“I think in looking for widely available and easy markers to obtain, CRP (C-reactive protein, a very good inflammatory marker) and albumin (common as a nutritional and acute phase marker) would be useful to add to this list.”

Unexplained weight loss is a nonspecific symptom that could indicate cancer. Nonspecific symptoms can result in delays in referrals and diagnoses.

“Unexplained weight loss is a good trigger for tracking blood test trends because it carries a real cancer risk (about 5% in this study), it doesn’t point to any one organ, and patients with it usually already have past blood results to look back on,” Thanki told MNT.

He added, “The best-supported other candidates are new-onset diabetes with weight loss in the elderly, which raises concern for pancreatic cancer, and a slow fall in Hb [hemoglobin] and MCV [mean corpuscular volume], which suggests colorectal cancer.”

However, Thanki added, “we need to keep in mind that the more widely you apply this monitoring, the more false alarms and unnecessary investigations you get. Sticking with the strongest relationships will hopefully prevent false alarms and unnecessary spending.”

The study authors note that the data available for their analysis was “limited to when patients had blood tests in clinical practice and to patients who had blood tests.”

Data was only analyzed from patients who had at least two tests in each trend window.

If a patient presents with unexplained weight loss but they had not had any blood tests during the relevant period beforehand, then there would be no way to assess historic trends when they first present with unexplained weight loss.

However, the authors also note that they found trend measurements to be most accurate when analyzing blood tests from within the last 1 year.

Therefore, even if a patient does not have any previous recent blood tests, the study findings suggest that it could still be beneficial to monitor trends in any additional blood tests going forward, particularly within the next 12-month period.

They also mention that not considering whether certain groups were more likely to be tested than others was a limit in the study’s approach.

“External validation and demonstrating a significant improvement clinically will be the biggest hurdles to making this into something that we can use in clinical practice,” concluded Thanki.