Validation of the Hair as a Tissue to Biomonitor PhIP, a Carcinogenic Heterocyclic Aromatic Amine

  • Topic: Cancer-related outcomes
  • Institution: University of Hawaii
  • Country: United States
  • Status: Completed

Scientific abstract

(View plain language abstract)

Aim

The purpose of this study was to validate the use of hair as a tissue in which to biomonitor chronic exposure to the carcinogen, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), a heterocyclic aromatic amine (HAA) formed in cooked meat. Short-term urinary biomarkers of PhIP exist; however, these are transient and only capture the preceding 24 hours of exposure. For individuals who chronically but intermittently consume cooked meats, urinary PhIP can be at undetectable levels, and these individuals can be misclassified. A biomarker of long-term exposure is critically needed to advance the field and confirm the carcinogenic effect of PhIP in humans.

Method & Results

With this WCRF grant, we further characterized our recently developed method to measure PhIP in non-dyed hair. We demonstrated the dose-sensitivity of the biomarker in a feeding study using two doses of PhIP intake and described the epidemiologic correlates of PhIP hair levels in a subset of 145 non-hair dye users in a case-control study of colorectal adenoma. In the feeding study, we showed a strong response to the intervention with a marked dose-dependent increase in hair PhIP levels normalized for hair melanin content, compared to baseline. This increase was linearly related to the dose received (r=0.55, p<0.0001). CYP1A2 activity, as measured by the caffeine test, and hair colour, did not modify the response to the intervention. The adenoma study showed that BMI and intake of processed meat and red meat cooked well-done with high temperature methods (grilled/barbecued, pan-fried, or oven-broiled), as assessed by a diet questionnaire, were significantly associated with PhIP hair levels. However, these questionnaire variables were only moderately good predictors of PhIP hair levels R2=0.2). Thus, the dietary questionnaire has only limited validity in measuring dietary exposure to HAAs.

Since hair dye is used by a large proportion of the population and because hair dye interferes with the measurement of PhIP by low-resolution triple quadrupole MS/MS assay, we also developed a new multistage, high-resolution accurate mass (HRMS) UPLC-MS-MS assay to quantify PhIP in dyed hair. This assay showed excellent validity and reproducibility. We demonstrated in a second feeding study that the increase in PhIP levels measured in dyed hair with this method in response to the intervention was similar to that obtained in subjects with non-dyed hair fed a comparable dose.

Conclusion

These studies establish the reliability of hair PhIP level as a biomarker of dietary exposure, and the practicality of using it in large epidemiological studies to study the association of HAAs with disease risk.

Plain language abstract

Background

The purpose of this study was to validate the use of hair as a tissue in which to assess long-term human exposure to the cancer-causing chemical, 2-amino-1-methyl-6-phenylimidazo[4,5-b]pyridine (PhIP), a heterocyclic aromatic amine (HAA) formed in cooked meat. Short-term urinary biomarkers of PhIP exist; however, these are short-lived and only capture the preceding 24 hours of exposure. For individuals who chronically but intermittently consume cooked meats, urinary PhIP can be at undetectable levels. A biomarker of long-term exposure is critically needed to advance the field and confirm the cancer-causing effect of PhIP in humans.

Aims & Objectives

We recently developed a convenient method to measure PhIP in hair, the levels of which varied greatly among subjects. In this WCRF-supported study we examined how dietary exposure to PhIP, the activity of a liver metabolic enzyme (CYP1A2) and hair colour affect hair PhIP level. Our aims were to: 1) validate the biomarker in a feeding study using already established protocols; and 2) describe its epidemiologic characteristics in a completed study of colorectal adenoma. Our objective was to determine the reliability of hair PhIP level as a biomarker of dietary intake, and the practicality of using it in large epidemiological studies to study the association of HAAs with disease risk.

How the study was carried out

Feeding study: Forty-one non-smoking volunteers were randomized to one of two doses of ground beef 
containing either 0.37 or 1.68μg/serving of PhIP, five days a week for a month, and then crossed over to the other dose after a wash-out period of 8-36 weeks. During the feeding phases, subjects ate lunch, five days a week, on the UH Manoa campus. Subjects were fed a standardized meal and refrained from eating any phytochemicals known to induce CYP1A2. For snacks, breakfasts and dinners, and on weekends, subjects followed their usual diet, except that they refrained from eating well-done meat/fish. During the wash-out period, the volunteers went back to their regular diet. The study was run in groups of ~10 subjects.

Adenoma Study: We investigated the epidemiologic correlates of hair PhIP levels in a completed study of colorectal adenoma in Hawaii in which a hair sample was collected from 229 participants, and in which smoking, hair-dye use and dietary intake, including HAA intake, were assessed through detailed questionnaires, and CYP1A2 activity measured with the caffeine test. A subset of 145 reported not using hair dye. All subjects were recruited among patients of an HMO (Kaiser Permanente Hawaii) who underwent a colonoscopy. Cases are those with a pathologically-confirmed adenoma; controls are those with a normal colon. Subjects provided clippings during a hair cut in the last year of the study.

Hair PhIP Analysis: Newly grown hair (50 mg) at the nape of the neck was clipped and finely minced with professional hair clippers. PhIP was measured by a low-resolution triple quadrupole MS/MS assay.

Key findings & conclusions

In the feeding study, we showed a strong response to the intervention with a marked dose-dependent increase in hair PhIP levels, compared to baseline. This increase was linearly and strongly related to the dose received. CYP1A2 activity, as measured by the caffeine test, and hair colour, after normalising for hair melanin content, did not modify the response to the intervention. 
The adenoma study showed that BMI and intake of processed meat and red meat cooked well-done by either grilling/barbecuing, pan-frying, or oven-broiling, as assessed by a diet questionnaire, were significantly associated with PhIP levels. However, these questionnaire variables were only moderately good predictors of PhIP hair levels (R2=0.2). Thus, the dietary questionnaire has only limited validity in measuring dietary exposure to HAAs. 
Since hair dye is used by a large proportion of the population and because it interferes with the measurement of PhIP, we also developed a new multistage, high-resolution accurate mass UPLC-MS-MS assay to quantify PhIP in dyed hair. This assay showed excellent validity and reproducibility. We demonstrated in a second feeding study that the increase in PhIP levels measured in dyed hair with this assay in response to the intervention was similar to that obtained in subjects with non-dyed hair fed a comparable dose. 
These studies established the reliability of hair PhIP level as a biomarker of dietary exposure, and the practicality of using it in large epidemiological studies to investigate the association of HAAs with disease risk.

Grant publications