Carcinogenesis Advance Access originally published online on January 24, 2008
Carcinogenesis 2008 29(9):1851; doi:10.1093/carcin/bgn022
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Reply: Letter to the editor in response to Morton et al. (2007) Carcinogenesis, 28, 1759–1764
The Personal Care Products Council, Washington, DC, USA
1 The Procter & Gamble Company, Cincinnati, OH, USA
* To whom correspondence should be addressed. Tel: +202 331 1770; Fax: +202 331 1969;Email: baileyj{at}personalcarecouncil.org
We read the recent article by Morton et al. (2007) with great interest (1). In this article, the authors examined non-Hodgkin lymphoma (NHL) risk in relation to reported hair dye use in a large US population-based case–control study. No overall association between reported use of hair dyes and NHL was found for either women [odds ratio (OR) = 1.2, 95% confidence interval (CI) = 0.9–1.6)] or men (OR = 0.9, 95% CI = 0.6–1.2). Analysis by type of product also showed no association between use of permanent hair dyes and NHL in either women (OR = 1.1, 95% CI = 0.8–1.6) or men (OR = 1.1, 95% CI = 0.7–1.8). The authors indicate that they collected data on color (e.g. black, brown, blonde, red) and what they referred to as ‘tone’ (light, dark) (more correctly referred to as ‘depth’ of color). In this publication, the authors defined the term ‘intense tone’ as black, dark brown and dark blonde shades, and they stated that intense tone products have higher concentrations of dye ingredients than ‘light or intermediate tone’ products. The authors cite a personal communication with J.A. Skare as support for these statements. As a point of clarification, Skare did not use the term intense tone in communications with the authors and did not make any statements that the concentration of dye ingredients in dark blonde permanent hair color products is similar to the concentrations found in black and dark brown permanent products. Black and dark brown products have higher concentrations of dyes in order to achieve the desired hair color, and a dark blonde product would have a much lower concentration of dye ingredients than black or dark brown products. Therefore, it is incorrect to conduct a stratified analysis based on hair dye color by combining black, dark brown and dark blonde as these authors have done.
In analyses of various stratified groups for evidence of dose–response relationships, no measure of association was statistically significant except for women who had used what the authors referred to as permanent intense tones for 
15 years beginning before 1980 (OR = 3.9, 95% CI = 1.2–12.5, based on 17 exposed cases and 4 exposed controls). However, the authors acknowledged that no consistent dose–response patterns were observed with frequency, duration or total lifetime applications among the users of permanent intense tone hair dyes. It is interesting to note that the analysis stratified for use of what was referred to as "permanent dark colors (black, brown, red)" for 15 years beginning before 1980 resulted in an OR = 1.5 (95% CI = 0.7–3.2, based on 23 exposed cases and 15 exposed controls). Clarification of specifically what colors (e.g. only dark browns and dark reds or all brown and reds, in addition to black) were included in this stratum and how that compares with the stratum defined as intense tones would be helpful.
The authors indicated that detailed hair coloring product-use histories were collected based on methods developed by Johns Hopkins University and Clairol, and they cited a final report to Clairol (ref. 51 cited by Morton et al.). Clairol has funded work to develop questionnaires for collecting data on past use of hair dyes, and this work has also been previously presented at a meeting of the American College of Epidemiology (2). In this project, structured questions were developed to collect detailed information on such parameters as type of product used (permanent, semi-permanent, temporary), color used (black, brown, blonde, red/auburn) and frequency and duration of use. These are important dimensions of hair dye exposure, and they have been used as criteria for evaluation of the quality of hair dye exposure assessment for published epidemiology studies in a recent review of the literature (3).
In reference to the collection of data on color, the use of a hair color chart is helpful as a visual aid for the respondent. Such a chart would typically include pictures that represent the full range of color categories available for permanent hair dyes, including black, brown (dark, medium, light), red (dark, medium, light) and blonde (dark, medium, light). The total concentration of dye ingredients present in hair dye products falling into these various categories of color will be dependent upon the particular brand of hair dye used, and each category of color will include products that exhibit a range of possible total dye concentrations. Some overlap in the ranges of total dye concentration across the spectrum of color categories can also be expected. However, the pattern for any given brand will be consistent in exhibiting the highest total dye concentrations in black color permanent hair dyes and the lowest concentrations in light blonde permanent hair dyes. When data on hair dye use are collected in studies, there will likely be some misclassification of color used due to inaccuracies in respondents’ reporting. Therefore, we suggest that information on color be collected using the 10 categories described above and that data on colors be grouped together into four strata for analysis: (i) light blonde; (ii) medium and dark blonde, light brown and light red; (iii) medium and dark brown, medium and dark red and (iv) black. These strata more appropriately group shades according to total dye concentrations and therefore represent a more appropriate approach to the categorization of color information than that conducted by Morton et al. in their analysis by intense tones.
Morton et al. also mentioned that hair dye formulations have been shown to contain 4-aminobiphenyl (4-ABP), although they indicated that 4-ABP has not been linked to lymphoma risk. With respect to bladder cancer risk, the trace quantities of 4-ABP reported by Turesky et al. (2003) in some hair dye formulations have been addressed by the Scientific Committee on Cosmetic Products and Non-Food Products Intended for Consumers (SCCNFP) in the European Union via a risk assessment issued in 2004 (4,5). In their assessment, the SCCNFP concluded that ‘even if "worst case" calculations are performed the amounts of 4-ABP reported in commercial hair dyes will not represent any risk of urinary bladder cancer in the users'. We agree with this conclusion.
We offer these comments in the spirit of maintaining open dialog with epidemiologists involved in conducting studies evaluating hair dye use and various types of cancer in order to improve upon the analysis and interpretation of results from such studies.
 |
Acknowledgments |
|---|
Conflict of Interest Statement: None declared.
|
References |
|---|
 Top References |
|---|
- Morton LM, et al. Hair dye use, genetic variation in N-acetyltransferase 1 (NAT1) and 2 (NAT2), and risk of non-Hodgkin lymphoma. Carcinogenesis. (2007) 28:1759–1764.
[Abstract/Free Full Text] - Pinney SM, et al. Hair dye use: choosing best questions and best method using confidence, concordance and reliability. Ann. Epidemiol. (2003) 13:572–573.
- Rollison DE, et al. Personal hair dye use and cancer: a systematic literature review and evaluation of exposure assessment in studies published since 1992. J. Toxicol. Environ. Health. Part B. (2006) 9:413–439.[CrossRef]
- Turesky RJ, et al. Identification of aminobiphenyl derivatives in commercial hair dyes. Chem. Res. Toxicol (2003) 16:1162–1173.[CrossRef][Web of Science][Medline]
- SCCNFP. Opinion Concerning Use of Permanent Hair Dyes and Bladder Cancer—Updated 2004 (2004) SCCNFP/0797/04. http://ec.europa.eu/health/ph_risk/committees/sccp/sccp_opinions_en.htm (23 April 2004).
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||