A routine blood test may pack a bigger punch than previously suspected, suggests a recent analysis of over 3 million Canadian patient records.
A finding of thrombocytosis (platelet count >450 x 109/L) was associated with a greatly increased risk for some cancers up to 5 years later.
Overall, a high platelet count increased by 2.7 times the odds of receiving a solid-tumor cancer diagnosis within 2 years (95% confidence interval, 2.6-2.8).
The cancers most likely to be associated with unexpected thrombocytosis were those notorious for late-stage diagnosis due to a lack of early symptoms.
The risk was highest (23.3 times) for ovarian cancer. The risk was 3.8 times higher for pancreatic cancer and 3.5 times higher for cervical cancer.
Lung cancer was 4.4 times more likely within 2 years among patients with thrombocytosis compared to patients with normal platelet counts.
Conversely, breast, prostate, and thyroid cancers were not linked to the finding of thrombocytosis.
The study results were published online in JAMA Network Open on Aug. 12).
One of the authors of the article, Stephen A. Narod, MD, director of the Familial Breast Cancer Research Unit at the Women’s College Research Institute, Toronto, said the results were not unexpected but “very striking.”
“I had a hunch we were going to see this because I’ve seen this in other databases,” said Dr. Narod. “I think what struck me about it was how ubiquitous it was.”
Dr. Narod urged physicians, especially those in primary care, to take note: “If the platelets are high, I would certainly have a concern about lung cancer, colon cancer, and ovarian cancer.”
Dr. Narod and coauthor Vasily Giannakeas, a PhD candidate, pointed out that in their analysis that they were unable to single out cases in which a blood test was performed because the patient complained of symptoms that are associated with cancer. In those cases, thrombocytosis may have been diagnostic, rather than a lifesaving serendipitous finding.
Similar findings were reported recently from the United Kingdom.
A study by Sarah Bailey, PhD, MPH, and colleagues that was published last year in the British Journal of General Practice also found a connection between cancer incidence and platelet count. Dr. Bailey is a senior research fellow at the University of Exeter, England.
However, unlike in the Canadian study, the team led by Dr. Bailey was able to distinguish those patients for whom there were alarm symptoms for cancer. Dr. Bailey and colleagues found that two-thirds of men older than 65 had “no recorded alarm features of cancer in the 21 days before their index platelet count.”
Although this suggests that a routine finding of thrombocytosis could uncover unsuspected cancers, Dr. Bailey is cautious about hailing platelet counts as a new cancer-screening tool.
In emailed comments, Dr. Bailey said, “The crucial part of our study is that it was conducted with patients who were ill enough to see their GP [general practitioner]. Opportunistic measurement in patients who are asymptomatic would be quite a different thing. We would have to study the platelet count and subsequent cancers in asymptomatic patients to know if that was worth doing.”
Perhaps most helpfully, the U.K. study showed that cancer risk was increased even among some patients with normal platelet counts. For example, for men aged 60 and older, lung cancer was 4.7 times more likely among those with high-normal counts (≥326 x 109/L).
Because of this somewhat alarming finding, Dr. Bailey suggested moving away from a focus on absolute values. Rising platelet counts might be more clinically useful, she said.
“Physicians should be on the lookout for any unexplained increase in an individual’s platelet count, irrespective of whether the increased value is over or under the local threshold that is applied to define thrombocytosis,” concluded Dr. Bailey.
Dr. Narod has disclosed no relevant financial relationships. Dr. Bailey is a research fellow of the CanTest Collaborative.
A version of this article first appeared on.