How are healthy biomarker ranges defined?

What makes a biomarker a biomarker? Testing for something doesn’t automatically make it a biomarker. A biomarker is the measurement or detection of a variable element of the human body. Imaging can yield biomarkers, as can environmental data. But these still have in common with physiological biomarkers the requirement for interpretation within a defined set of values, or reference ranges.

Reference ranges can be the topic of much debate; an analytical assay will have a continuum of output values, along which we apply gateway posts defining “High”, “Low”, and “Normal”. Even where a pathogen may be simply “Detected” or “Undetected”, the cycle number of qPCR reaction or reactivity level of a serology test are numerically defined to delineate presence or absence.

In theory, a reference range should be a universal physiological constant that applies to humans regardless of where or how the test is performed. Diagnostic tests should be assessed for accurate and reliable measurement of true analyte values. However, we’ll explore why defining a reference range is often not so straightforward.

Reference ranges are defined by study populations

The healthy or “Normal” reference range for a biomarker is commonly defined as covering 95% of a population. This is often defined by the manufacturers of the analytical assay like Roche or Siemens, submitted as part of certification under regulatory frameworks, and included in the assay Instructions for Use. So, values in the bottom 2.5th percentile and above the 97.5th percentile of a population distribution will be classified as “Low” or “High”, respectively. Some biomarkers may not have a lower limit below which there is a risk to health, so they will be either “Normal” from zero up to a certain value above which they are classed as “High”.

For example, alanine aminotransferase (ALT) is a liver enzyme that is released into the blood when your liver is damaged or under stress. It can be used to help diagnose some liver conditions, as it is normally not found in elevated levels in the blood. The upper bounds of a “Normal” result is commonly defined around <50 IU/L in males, and <35 IU/L in females.

Populations can be updated with new data

Reference interval ranges are referred to in ISO15189:2022 – the regulatory compliance all medical laboratories must achieve and adhere to, but there is no stipulation that laboratories must adhere to one particular source of reference ranges.

The ranges supplied by the assay manufacturer are based on a sufficient population size to achieve accreditation. This can sometimes be 100 or fewer individuals. Pathology labs running routine blood tests will often instigate, or be involved in, additional studies involving a much higher N and more diverse population sets. If a study is performed that yields an altered set of ranges the lab may alter the range limits set into its laboratory information management system (LIMS), and issue a change notification email to all of its customers.

Does genetic background matter?

Some variation may exist across populations with different genetic backgrounds and ethnicity. However, there is some debate over the validity of applying broad stroke changes to healthy ranges depending on someone’s ethnicity, especially where this may change a healthcare treatment path and risk perpetuation of racial stereotyping. In the UK, different ranges for different ethnicities are generally not applied, however, as with all diagnostics doctors take into consideration a patient’s overall health and family history.

Gender affects many biomarker reference ranges

Many biomarkers vary significantly in normal healthy levels between males and females, including ALT above. Most significant are, of course, the sex hormones like oestrogen, luteinising hormone (LH), follicle stimulating hormone (FSH), and progesterone that fluctuate throughout the menstrual cycle and play vital roles in onset of puberty, fertility, and menopause. Precise levels can be tracked in detail, but the interpretation cannot be easily assigned automatically, so reference ranges are often provided separately to allow the patient to interpret results in a wider context not visible to the lab.

Age, particularly early development, can influence biomarker ranges

Age is a significant factor affecting many biomarkers, in older age, but especially in early development. This can be linked to many things, including hormonal development or increasing muscle mass. Diagnostics are, of course, an essential part of paediatric care pathways and need to be interpreted with care and caution. Because of the variability in healthy reference ranges throughout each child’s development, there is a high risk of erroneous interpretation, so many diagnostics companies put age restrictions on home testing.

If you are looking for testing for your child or are a minor, please consult your doctor and check for any age restrictions.

Patient context matter above all

In all cases, biomarker results should be read with as much context as possible and not in isolation.

Have you recently had a meal? Have you been exercising? Did you recently have a cold? Many factors can affect biomarker results.

So, who decides on reference ranges?

We’ve established some of the main reasons reference ranges can vary, so where does the authority lie on what we should be using?

In the UK, the Royal College of Physicians issues guidelines on acceptable biomarker results interpretation. However, NHS trusts around the country also issue their own guidelines. In the US, hospital systems publish biomarker information and guidelines.

For most diagnostic testing solutions, the laboratory is the nexus, where primary result information is produced and the interpretation of the results applied according to reference ranges internalised in the LIMS.

There has been some debate over the need to harmonise reference ranges and even include more stringent guidelines as part of ISO15189 certification. Harmonisation has occurred for high risk biomarkers like potassium, but proponents are advocating for a broader range of standards across other biomarkers as well.

About the author

Toby Call is co-founder of research and innovation at Hurdle. He has a background in synthetic biology and a PhD in industrial biotechnology and biochemistry. He works with operations, product, science, and medical teams to bring biomarker solutions to life in R&D and diagnostic settings.