This article presents the results of analyses of hair fibers from women with breast cancer using synchrotron-derived X-ray diffraction. These diffraction patterns contained a new feature superimposed on the normal diffraction pattern of a-keratin. The feature appeared as a ring with a molecular spacing determined to be 4.76 6 0.07 nm. This feature was not present in the diffraction patterns of hair from women without breast cancer as assessed by other routine clinical diagnostic techniques. Furthermore, different hairs from the same subject analysed on two different synchrotron beamlines give remarkably consistent diffraction patterns. Previous studies by other investigators have suggested that analysis of X-ray diffraction patterns of hair can reveal the presence of breast cancer in clinical and preclinical trials. This finding, however, has not been independently confirmed. The methodologies of sample handling, sample exposure and image analysis are known to be vital. We discuss some of these issues and provide a detailed description of the methodology employed for the sample handling and image analysis and new methodologies developed from this work. We conclude that X-ray diffraction of hair has the potential to provide a non-invasive test for the presence of breast cancer. ' 2007 Wiley-Liss, Inc.Key words: X-ray diffraction; hair; breast cancer; synchrotron In 1999 James et al. reported differences in the X-ray diffraction patterns of hair from individuals with breast cancer compared to healthy subjects.1 The X-ray diffraction patterns of hair from cancer patients contained a ring of comparatively low intensity, which was superimposed on the normal a-keratin pattern obtained from healthy control subjects and which corresponded to a molecular spacing of d 5 4.44 nm. This ring was reportedly observed in all samples of scalp and/or pubic hair taken from women diagnosed with breast cancer, as well as from subjects ''not yet diagnosed with breast cancer but suspected of being at risk''. In other words, a number of false positives were identified. Subsequent papers by James and coworkers reported X-ray diffraction analysis results of blinded human samples, which were consistent with the initial publication.2,3 The later paper reported on the results of 503 blinded hair sample analyses and demonstrated a sensitivity of 100% (no false negatives) and a specificity of 86% (14% false positives by comparison to mammography) for breast cancer.However, the finding remained highly controversial as several groups independent of James attempted to replicate the original findings and all were unsuccessful. [4][5][6][7][8][9][10][11] The technical explanations for these replication failures [12][13][14][15][16] provide insight into the experimental challenges. Myer et al., 9 whose diffraction data was not of sufficient quality to demonstrate a correlation between an altered X-ray diffraction pattern and the presence of breast cancer, sent the 27 samples they had examined to James who subsequently performed X-ray diffraction a...