Correspondence re: L. R. Kidd et al., Urinary Excretion of 2-Amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP) in White, African-American, and Asian-American Men in Los Angeles County. Cancer Epidemiol. Biomark. Prev., 8: 439–445, 1999

Dietary meat intake has been associated with an increase of cancer risk for a number of organ sites including colon, breast, and pancreas. Recent studies point towards the possible role of HAAs²formed upon heating of meat. These have been shown to be genotoxic in vitro and carcinogenic in rodents albeit at rather high concentrations.

To estimate the role of these HAAs, Kidd et al.(1) and others (2, 3, 4) published biomonitoring studies evaluating the quantitative analysis of urinary levels of HAAs or their respective metabolites as a measure of the intake of HAAs. A more indirect approach has been the measurement of urinary mutagenic activity (5, 6).

As for most urinary biomarkers in toxicology and pharmacology, the measurement of absolute values in (repeated) 24 h urine samples appears to be the most reliable method. It has been shown (2, 4) that renal excretion of HAAs is complete with 12 h after ingestion.

Because collection of 24 or 12 h urinary samples is difficult, and these are often unavailable, normalizing of urine for creatinine has been used to allow comparison of individuals, thus excluding variations in urinary flow. Levels of creatinine excretion are regarded as a measure of renal function, and creatine catabolism and creatinine clearance are relatively constant (7).

Kidd et al. (1) analyzed urinary samples from different ethnic groups and were unable to observe a correlation of data from a dietary questionnaire with urinary levels of 2-amino-1-methyl-6-phenylimidazol[4,5-b]pyridine normalized for creatinine.

However, considerable amounts of creatinine are ingested with a meal of meat or fish that may contain 2–4 mg creatine/g wet weight, with creatinine being produced from creatine when meat is cooked. A postprandial increase in creatinine plasma levels of up to 65% is observed. Similarly, urinary creatinine levels are dependent on dietary intake of meat (8, 9).

Furthermore, creatinine has been shown to be a precursor of the formation of mutagenic HAAs with an aminoimidazo moiety such as 2-amino-3-methyl-3H-imidazo[4,5-f]quinoline,2-amino-3,8-dimethyl-3H-imidazo[4,5-f]quinoline, or 2-amino-1-methyl-6-phenylimidazol[4,5-b]pyridine. Indeed,HAA formation is dependent on creatinine concentrations in model systems (10). It has been shown that HAA levels in fried meat increase with cooking time and cooking temperature (11).

Therefore, systematic errors would be introduced in a study comparing patients with low intake of heavily cooked meat and patients with high intake of lightly browned meat: both groups may have similar HAA intake. If urinary levels of HAAs (and metabolites) or excreted mutagenicity is considered on a 24-h basis, both would be similar. If HAA levels are normalized for creatinine, patients consuming high amounts of lightly cooked meat (and thus with higher urinary creatinine levels) will appear to have a lower intake of HAA.

This effect may be even more pronounced when limited urine samples are collected, which may be within a few hours after a meat meal, rather than 24 h urine samples.

Similarly, in a study comparing urinary levels of HAAs (or mutagenicity) before and after a meal of fried meat, normalizing for creatinine content will diminish a possible postprandial increase of these biomarkers.

Therefore, one has to be cautious to normalize urinary biomarkers to creatinine levels in urine. It appears to be inappropriate to normalize excretion of HAAs to creatinine levels because this calculation may lead to spuriously low values.