A patient’s blood test history may reveal more about future cancer risk than any single result, according to research published in PLOS Medicine. The study suggests that gradual changes in routine blood tests can begin several years before cancer becomes clinically apparent, particularly in people who later also develop unexplained weight loss.
The findings could strengthen cancer risk assessment in primary care. At present, clinicians often judge blood tests by asking whether an individual result falls outside a laboratory’s reference range.
“They found that blood test patterns associated with cancer often began to emerge three to five years before the onset of unexplained weight loss”
The new research points towards a different approach, examining how a person’s results change over time, then interpreting those changes in the context of age and sex.
That shift could be important. A result labelled “normal” may still represent a meaningful change for one patient. Conversely, a mildly abnormal value may be less concerning in another person, depending on their age, medical history and previous results.
The study examined 275,205 patients who had presented with unexplained weight loss. Nearly 14,000 were diagnosed with cancer within six months of that presentation. Researchers analysed trends in 26 commonly performed blood tests, looking at results recorded one, three, six and 10 years before the weight loss was reported.
They found that blood test patterns associated with cancer often began to emerge three to five years before the onset of unexplained weight loss. In many cases, the results remained within the accepted reference range. This means that a pattern of gradual change might be overlooked when tests are reviewed individually.
The strongest associations involved markers linked with inflammation, blood loss, red blood-cell production and nutritional status. These included C-reactive protein, erythrocyte sedimentation rate, plasma viscosity, haemoglobin, haematocrit and red cell distribution width. Some findings varied according to age.
For example, changes in haemoglobin and red blood-cell measures were particularly relevant among people aged 70 and above. Changes in haematocrit and red cell distribution width were more notable among adults aged 50 and under. These differences underline the limitations of applying the same interpretation to every patient.
The researchers also adjusted their analysis according to age and sex. This improved the ability of the blood test trends to distinguish people who later received a cancer diagnosis from those who did not. The highest reported area under the curve was 0.82. In diagnostic research, a score of 1.0 represents perfect discrimination, while 0.5 indicates performance no better than chance.
The result does not mean that blood test trends can diagnose cancer. It indicates that they may provide useful additional information when combined with symptoms, examination findings and other clinical factors.
The findings support a more personalised approach to interpreting laboratory results. Rather than comparing a patient only with a population-based reference interval, clinicians could also compare the patient with their own previous results.
This is particularly relevant when unexplained weight loss is present. Weight loss can result from many conditions, including digestive disorders, infection, hormonal disease, mental health problems and medication effects. It is not specific to cancer. Yet the study found that around 5% of patients with unexplained weight loss received a cancer diagnosis within six months, making the symptom an important warning sign.
A falling haemoglobin level may suggest ongoing blood loss or another underlying process. A gradual reduction in mean cell volume can sometimes point towards iron deficiency, including blood loss from the bowel.
New-onset diabetes accompanied by weight loss in an older adult may also raise concern about pancreatic disease, although most people with these findings will not have pancreatic cancer.
The importance lies in the combination of evidence. A small change in one blood test may be harmless. Several related changes, developing over time and occurring alongside weight loss or another concerning symptom, may warrant further assessment.
The study identified relationships between particular blood test patterns and certain cancer types. Trends in white blood cell counts and neutrophils were associated with lung cancer. Changes in mean cell volume were linked with bowel cancer and lymphoma. Trends involving red blood cell count, aspartate aminotransferase, haematocrit and the platelet-to-lymphocyte ratio were associated with prostate cancer.
These links should be interpreted carefully. None of the blood tests can confirm a specific cancer on its own. Many conditions can alter inflammation markers, blood counts or liver enzymes. Infection, autoimmune disease, kidney problems, nutritional deficiencies and medicines may all produce similar changes.
The value of the approach is therefore likely to be one of risk assessment rather than diagnosis. A computer system or laboratory report might identify a notable change, allowing a clinician to review the patient’s symptoms and decide whether further tests, imaging or referral are appropriate.
The most commonly available tests in the study were haemoglobin, white blood cell count and platelets. These tests are widely used, relatively inexpensive and often included in routine blood counts. They can provide clues about inflammation, blood loss, infection and nutritional problems.
Other markers could potentially add useful information. Experts have highlighted C-reactive protein as a practical indicator of inflammation. Albumin may also be relevant because it can reflect nutritional status and the body’s response to illness. Further research will be needed to determine which combination of tests provides the greatest benefit without producing too many false alarms.
That balance is crucial. The broader the screening approach, the more likely it is to identify harmless fluctuations. False-positive results can lead to anxiety, repeat testing, scans, invasive procedures and unnecessary healthcare costs. A useful system must distinguish clinically meaningful patterns from ordinary biological variation.
The study also has important limitations. The analysis included only people who had undergone blood testing in routine clinical practice. Patients with fewer tests, less contact with healthcare services or different patterns of access may have been under-represented.
Participants needed to have at least two blood tests in each period used to assess a trend. This excluded people without sufficient historical data. Someone presenting with unexplained weight loss may have no comparable results from previous years, making it impossible to calculate a long-term trajectory.
The researchers found that trends measured using blood tests from the previous year were often the most accurate. This suggests that long-term records are not essential in every case. If earlier results are unavailable, monitoring new results over the next 12 months may still provide useful information, particularly when symptoms persist or develop.
Another limitation was the inability to fully account for why some patients were tested more frequently than others. People with chronic illnesses, older adults and those who have regular contact with medical services may have richer laboratory histories. Their results could appear easier to interpret simply because more data are available.
The next step is external validation. The findings need to be tested in different healthcare systems and populations. Researchers must also determine whether using blood test trajectories actually leads to earlier diagnosis, faster referral or better outcomes. A statistically strong prediction model is not automatically a clinically useful one.
Implementation would require careful design. Laboratory systems could compare a new result with earlier values, highlight meaningful changes and adjust interpretation according to age and sex. Alerts would need to be specific enough to avoid overwhelming general practitioners with warnings.
Clinical judgement would remain central. Blood test trends should support, not replace, a discussion with the patient. Symptoms, family history, smoking status, medications, existing conditions and examination findings all affect cancer risk.
For patients, the message is not to interpret every fluctuation as a sign of cancer. Blood results naturally vary. A single abnormal value often has a benign explanation, particularly if it returns to normal. However, persistent or progressive changes deserve attention, especially when accompanied by unexplained weight loss, persistent pain, altered bowel habits, unusual bleeding, ongoing cough or marked fatigue.
Patients can help by telling their healthcare professional about changes in weight, appetite, bowel or bladder habits and general wellbeing. Keeping track of previous test results may also help clinicians understand whether a value is new, longstanding or gradually changing.
The research presents a simple but potentially powerful idea, the direction of a blood test may matter as much as its position. A snapshot can miss a slow biological shift. A sequence may reveal it.
More work is needed before this method becomes standard practice. Even so, the study supports a growing interest in using routine healthcare data more intelligently. Blood tests already collected for many other reasons could, when interpreted over time, offer an additional opportunity to recognise cancer earlier.
The post Unexplained Weight Loss and Changing Blood Test Trends May Offer Strong Early Warning Signs of Cancer first appeared on PP Health Malaysia (PPHM) | 10POINTS Evidence-Based Health News.
