Miriam E. Tucker, for Medscape

Gerry Gajadharsingh writes:

“More research confirming that blood test results, even in the so-called normal medical range can be associated with an increased risk of disease. In this case, the researchers looking at the UK biobank data looked at the relationship between blood glucose and increasing lung cancer risk. Most clinicians tend to use a marker called HbA1c, a red blood cell tends to live for around 120 days and labs can measure the glycated portion (sugar portion) of the red blood cell, essentially giving us a blood glucose marker over a three-month period.

 The important point was the relationship between incident lung cancer and HbA1c was nonlinear with the strongest associations occurring at A1c levels of 32-42 mmol/mol (5.1%-6.0%), please note that this is still in the normal medical range and levels that fall short of the cut-off for a diagnosis of diabetes, and at the lower end, fall short of the definition of prediabetes.

 Researchers subdivided the study cohort based on 2009 criteria for diabetes and prediabetes. They classified those with an A1c of less than 42 mmol/mol (less than 6.0%) as having normoglycemia; about 91% of the cohort. They classified those with an A1c of 42-47 mmol/mol (6.0%-6.4%) as having prediabetes; about 3% of the cohort. And they classified those with an A1c of at least 48 mmol/mol (at least 6.5%), or self-reported a diagnosis of diabetes, or were taking antidiabetic medications, as having diabetes; about 6% of the cohort.

 Elevated A1c, prediabetes, and diabetes were each significantly associated with an increased incidence of lung cancer.

 But in an analysis that considered A1c as a continuous variable, and after adjustment for all potential confounders, the incident lung cancer risk decreased with increasing levels of A1c until about 32 mmol/mol (5.1%), then increased sharply until around 42 mmol/mol (6.0%), and then increased at a more gradual rate as A1c levels further increased.

 Keeping blood glucose in optimal ranges, as opposed to standard medical ranges in my opinion is critically important for many health conditions. Until recently, there has not been much attention at research, focusing on functional ranges in blood tests. Let’s hope that the research keeps on building.”

 If you’re interested in finding out more about blood test, interpretation and functional ranges please go to

https://www.thehealthequation.co.uk/thehealthequationcourses/

 

 Key Takeaways

Results from a large database study has shown a positive association between prediabetes, diabetes, and incident lung cancer.

Treatment with glucose-lowering medications was significantly linked with a reduced rate of incident lung cancer.

The relationship between incident lung cancer and HbA1c was nonlinear with the strongest associations occurring at A1c levels of 32-42 mmol/mol (5.1%-6.0%), levels that fall short of the cut-off for a diagnosis of diabetes, and at the lower end, fall short of the definition of prediabetes.

The risk of lung cancer was also more pronounced among participants less than 60 years old, current smokers, and those with a high genetic risk for lung cancer.

Why This Matters

Lung cancer is the leading cause of cancer-related mortality worldwide.

Until this study, little has been known about the risk for incident lung cancer across the glycaemic spectrum. Previous population-based studies investigating the link between diabetes and prediabetes and lung cancer have had mixed results.

The findings are new evidence that prediabetes and diabetes increase the risk of incident lung cancer and that use of antidiabetic medications may reduce the risk.

Future research into biologic mechanisms is needed to explain the association between blood glucose and lung cancer and to inform clinical practice and health policy.

Study Design

The study used data from a total of 283,257 cancer-free adults recruited between 2006 and 2010 in the UK Biobank.

Researchers subdivided the study cohort based on 2009 criteria for diabetes and prediabetes. They classified those with an A1c of less than 42 mmol/mol (less than 6.0%) as having normoglycemia; about 91% of the cohort. They classified those with an A1c of 42-47 mmol/mol (6.0%-6.4%) as having prediabetes; about 3% of the cohort. And they classified those with an A1c of at least 48 mmol/mol (at least 6.5%), or self-reported a diagnosis of diabetes, or were taking antidiabetic medications, as having diabetes; about 6% of the cohort.

Key Results

Among the 283,257 participants, 2355 (0.83%) developed lung cancer during a median follow-up of about 11 years.

Elevated A1c, prediabetes, and diabetes were each significantly associated with an increased incidence of lung cancer.

After adjustment for all potential confounders, each standard deviation (6.67 mmol/mol) increase in A1c was associated with a significant 3% increase in the risk of incident lung cancer.

Those with prediabetes had a significant 37% increased rate of lung cancer compared with people with normoglycemia after full adjustment, and people with diabetes had a significant 20% relative increase.

Participants with diabetes who were taking glucose-lowering medication did not have a significantly increased risk of incident lung cancer, but people with diabetes and not taking any medication had a significant 33% increased incidence of lung cancer.

In an analysis that considered A1c as a continuous variable, and after adjustment for all potential confounders, the incident lung cancer risk decreased with increasing levels of A1c until about 32 mmol/mol (5.1%), then increased sharply until around 42 mmol/mol (6.0%), and then increased at a more gradual rate as A1c levels further increased.

The dose–response association between A1c and incident lung cancer was stable when stratified by age, sex, ethnicity, smoking status, and genetic risk.

Limitations

Only baseline data on glycaemic status were used.

A large number of participants were excluded because of missing data.

The findings may have been affected by unknown confounders.

Generalizability of the findings is uncertain beyond the predominantly White study population.