© The Author 2007. Published by Oxford University Press.
CORRESPONDENCE |
Re: Prostate Cancer in Fathers With Fewer Male Offspring: the Jerusalem Perinatal Study Cohort
Correspondence to: William H. James, PhD, The Galton Laboratory, University College London, Wolfson House, 4 Stephenson Way, London NW 1 2 HE, UK (e-mail: whjames{at}waitrose.com).
Harlap et al. (1) reported that the sex ratio (proportion male) of offspring of their sample of Jewish Israeli men with prostate cancer was statistically lower than that of control men without cancer or than that of control men with other forms of cancers. These authors write that "Previous studies of this association [between prostate cancer and offspring sex ratio] have yielded conflicting results’. They cited three previous (Canadian) studies (2–4), which presented such data. These data [together with those of Harlap et al. (1)] are summarized in Table 1. For completeness, I have added two further small sets of data (5) on white and black Californians.
|
Harlap et al. (1) are correct about the conflicting results: the heterogeneity chi-square across the three Canadian studies takes the value of 7.8, with 2 df and P value less than .02.
However, they do not address the contrasts between these previous datasets and their own. If the previous Canadian data (2–4) are all pooled and tested against those of Harlap et al. (1), 
2 = 8.6 with 1 df, P value less than .005. Moreover, the heterogeneity chi-square calculated across these four samples takes a value of 16.5 with 3 df, P value less than .001. Lastly, a heterogeneity chi-square value calculated from all six datasets in the Table takes a value of 17.1 with 5 df, P value less than .005.
Harlap et al. (1) offer a hypothesis to explain their low sex ratio. But a more immediate problem is presented by this heterogeneity. Before trying to explain one of these sets of data, we need to know the cause of this heterogeneity among the sets. I should declare an interest here. In 1990, I interpreted the then available data (which showed a statistically significant increase in the ratio of males to females) as being consistent with the notion that prostate cancer and the sex ratio were both caused by elevated androgen levels (5). In the face of the heterogeneity, that interpretation is no longer tenable without qualification.
REFERENCES
(1) Harlap S, Paltiel O, Friedlander Y, Calderon-Margalit R, Deutsch L, Kleinhais KR, et al. Prostate cancer in fathers with fewer male offspring: the Jerusalem perinatal study cohort. J Natl Cancer Inst (2007) 99:77–81.
(2) Hill GB, Fincham SM, Wijayasinghe C, Haronga CL, Hendin MM. Sex ratio of offspring of patients with prostatic cancer. Can Med Assoc J (1985) 133:567–71.[Abstract]
(3) Le Marchand L, Yoshizawa CN, Kolonel LN. Sex ratio of offspring of patients with prostatic cancer. Can Med Assoc J (1986) 135:107.[Medline]
(4) Spitz MR, Berman EL, Newell GR, von Eschenbacch AC. Sex ratio of offspring of patients with prostatic cancer. Can Med Assoc J (1986) 134:104–5.[Medline]
(5) James WH. The hypothesized hormonal control of human sex ratio at birth—an update. J Theor Biol (1990) 143:555–64.[Web of Science][Medline]
(6) Allan BB, Brant R, Seidel JE, Jarrell JF. Declining sex ratios in Canada. Can Med Assoc J (1997) 156:37–41.[Abstract]
Related Article in JNCI
Response to this Correspondence
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2007 99: 77-81.
J Natl Cancer Inst 2007 99: 903-904.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||