Joel Brind

Induced abortion as an independent risk factor for breast cancer: a comprehensive review and meta-analysis.

STUDY OBJECTIVE: To ascertain, from the published reports to date, whether or not a significantly increased risk of breast cancer is specifically attributable to a history of induced abortion, independent of spontaneous abortion and age at first full term pregnancy (or first live birth); to establish the relative magnitude of such risk increase as may be found, and to ascertain and quantify such risk increases as may pertain to particular subpopulations of women exposed to induced abortion; in particular, nulliparous women and parous women exposed before compared with after the first full term pregnancy. INCLUDED STUDIES: The meta-analysis includes all 28 published reports which include specific data on induced abortion and breast cancer incidence. Since some study data are presented in more than one report, the 28 reports were determined to constitute 23 independent studies. Overall induced abortion odds ratios and odds ratios for the different subpopulations were calculated using an average weighted according to the inverse of the variance. An overall unweighted average was also computed for comparison. No quality criteria were imposed, but a narrative review of all included studies is presented for the readers use in assessing the quality of individual studies. EXCLUDED STUDIES: All 33 published reports including data on abortion and breast cancer incidence but either pertaining only to spontaneous abortion or to abortion without specification as to whether it was induced or spontaneous. These studies are listed for the readers information. RESULTS: The overall odds ratio (for any induced abortion exposure; n = 21 studies) was 1.3 (95% confidence interval of 1.2, 1.4). For comparison, the unweighted overall odds ratio was 1.4 (1.3,1.6). The odds ratio for nulliparous women was 1.3 (1.0,1.6), that for abortion before the first term pregnancy in parous women was 1.5 (1.2,1.8), and that for abortion after the first term pregnancy was 1.3 (1.1,1.5). CONCLUSIONS: The results support the inclusion of induced abortion among significant independent risk factors for breast cancer, regardless of parity or timing of abortion relative to the first term pregnancy. Although the increase in risk was relatively low, the high incidence of both breast cancer and induced abortion suggest a substantial impact of thousands of excess cases per year currently, and a potentially much greater impact in the next century, as the first cohort of women exposed to legal induced abortion continues to age.

Elevated serum dehydroepiandrosterone sulfate levels in practitioners of the Transcendental Meditation (TM) and TM-Sidhi programs

Serum dehydroepiandosterone sulfate (DHEA-S) levels were measured in 270 men and 153 women who were experienced practitioners of the Transcendental Meditation (TM) and TM-Sidhi programs, mental techniques practiced twice daily, sitting quietly with the eyes closed. These were compared according to sex and 5-year age grouping to 799 male and 453 female nonmeditators. The mean DHEA-S levels in the TM group were higher in all 11 of the age groups measured in women and in 6 of 7 5-year age groups over 40 in men. There were no systematic differences in younger men. Simple regression using TM-group data revealed that this effect was independent of diet, body mass index, and exercise. The mean TM-group levels measured in all women and in the older men were generally comparable to those of nonmeditator groups 5 to 10 years younger. These findings suggest that some characteristics of TM practitioners are modifying the age-related deterioration in DHEA-S secretion by the adrenal cortex.

Enhanced Yield of Antihemophilic Factor and von Willebrand Factor by Cryoprecipitation with Polyethylene Glycol1

Abstract. A procedure involving cryoprecipitation of human plasma in the presence of polyethylene glycol results in consistently higher yields of antihemophilic factor and von Willebrand factor than were heretofore achieved by cryoprecipitation alone. The resulting AHF and vWF are at least as pure as ordinary cryoprecipitate and can easily be further purified by additional processing. The method is expected to be especially useful in the preparation of AHF and vWF concentrates for the treatment of patients with hemophilia or von Willebrands disease.

Abortion and breast cancer

Goldacre et al 1 use an impressively large database (28 616 cases and 325 456 controls) and an observed/expected ratio significantly below unity (0.83; 95% CI 0.74 to 0.93) in their record linkage study, to conclude that induced abortion “does not increase the risk of breast cancer.” In their discussion, however, the authors acknowledge a massive deficiency—that is, that their “data on abortions are substantially incomplete because they only include women admitted to hospital (and) only include those in the care of the National Health Service (NHS)”. Considering that the majority of English abortions do not occur in NHS hospitals, most of the women in the study who did indeed have an induced abortion are probably misclassified as not having had any. The even more egregious nature of this flaw is reflected in the fact that a mere 300 cases—just over 1% of the total—are classified as having had an induced abortion. As the overall induced abortion rate in England and Wales averaged more than 1% per year during the study period (1968–1998),2 it is conservatively estimated that approximately 15% of the women in the cohort underwent an induced abortion in their lifetime. Consequently, more than 90% of the women in the study cohort who underwent induced abortion were misclassified as not having an induced abortion. Therefore, the Goldacre et al dataset is wholly inapplicable to the question of an association between induced abortion and breast cancer. Such inappropriate use of a large dataset is reminiscent of a similar report from 1982 by an Oxford group with authorship overlapping that of the present study (D Yeates).3 In the 1982 study, Vessey et al , using …

Radioimmunoassay of estrone sulfate in the serum of normal men after a non-chromatographic procedure that eliminates interference from dehydroepiandrosterone sulfate

Duplicate aliquots of 20 fresh-frozen normal human male sera were prepared for estrone sulfate (ES) radioimmunoassay (RIA) by each of three different methods: the thin-layer chromatography (TLC) procedure we previously reported, a new procedure including overnight heating (100 C) of an ethanol extract reconstituted in dilute acetate buffer, and the new procedure with the hot incubation omitted. The purpose of the 100 C incubation was the selective thermal solvolysis of dehydroepiandrosterone sulfate (DS), the only steroid conjugate present in serum in high enough concentrations to interfere with a high-specificity ES RIA. Dehydroepiandrosterone released by solvolysis and endogenous unconjugated steroids were extracted from the samples with ether before RIA. Estrone sulfate values obtained after the thermal solvolysis preparation averaged 854 +/- 501 pg/ml (SD) versus 826 +/- 474 pg/ml (SD) after the TLC method, with excellent correlation between the two (r = 0.97). Samples prepared by the new method but with thermal solvolysis omitted averaged a 33.8% elevation of measured ES level, an elevation significantly correlated (P less than 0.02) with DS levels obtained from the same specimens. In addition, a single specimen showed no elevation after preparation by the thermal solvolysis method when up to 8 micrograms/ml authentic DS as added before extraction. Compared with the TLC method, the new method also provides substantial savings in specimen volume requirements and sample processing time.

Esterase activity in human breast cyst fluid : associations with steroid sulfates and cations

Human breast cyst fluid (BCF) contains an esterase that on the basis of electrophoretic mobility and response to inhibitors differs from those found in the plasma. From a total of 384 BCF samples analyzed for esterase using p-nitrophenyl hexanoate as substrate, 149 (39%) showed significant activity. The samples had been analyzed for the concentrations of the sulfates of estrone, estriol, dehydroepiandrosterone, as well as the potassium and sodium cations (K+/Na+). The data were submitted to statistical analysis using the Spearman rank order test. The esterase-positive samples exhibited a significant positive association with each of the steroid sulfates and the K+/Na+ ratios. Except for protein concentration, there was no significant correlation between the esterase-positive and esterase-negative cysts. These observations may have physiological significance in that high K+/Na+ ratio cysts have been related to the histological status of the cyst.

Radioimmunoassay of estrone sulfate in the serum of normal men after a chromatographic procedure that eliminates dehydroepiandrosterone sulfate interference

Fifty fresh-frozen normal male sera containing tritiated estrone sulfate (ES) and dehydroepiandrosterone sulfate (DS) were extracted with ethanol after ether extraction of unconjugated steroids. Washed extracts were defatted and chromatographed on polyamide-coated plates by reversed phase paired ion TLC. Plates were scanned for radioactivity, and ES peaks were cut, eluted and assayed by direct RIA with a commercially available antiserum. Mean ES values were 445 +/- 209 pg/mL (SD), in agreement with the three lowest of the seven laboratories which had previously reported normal male ES values. No differences were observed in ES values when samples were rechromatographed prior to assay, or when up to 4 micrograms/mL unlabeled DS was added to serum before extraction. These data confirm the absence of interference by DS in the current study and suggest that previously reported high (716-1194 pg/mL) mean normal male ES values reflect DS interference. The present study also demonstrates the the stability of ES in sera stored frozen at -40 C for an average of 17 years (mean: 406 +/- 258 pg/mL; [SD]; n = 41).

Breast cancer and induced abortions in China

Sir, A recent study in this journal (Ye et al, 2002) has reported a slight but nonsignificant elevation in risk among Chinese women who had reported any induced abortions. On this basis, the authors conclude: ‘Abortions as they have been performed in China are not an important cause of breast cancer.’ This finding appears to be strengthened by the fact that the same odds ratio (OR=1.06) was obtained on their cohort of Shanghai textile workers either when a cohort analysis was performed, or when an age-matched case–control study was drawn from the cohort. The essentially null association was additionally reinforced by a similar finding in a previously published study on Shanghai women (Sanderson et al, 2001). In China, it is well known that the governments ‘one child policy’ has led to a pattern in which induced abortion is very common, and almost always used after the first (usually, only) birth. Hence, women who are exposed to induced abortion would tend to be those who have their child(ren) at a younger age. The unexposed women, therefore, constitute a population that includes more women who are nulliparous or who had their child(ren) at a later age. Both of these characteristics are universally acknowledged risk factors for breast cancer, and were in fact observed by Ye et al in their cohort. The consequence of such population characteristics is confounding in the direction of underestimating the relative risk. That is, those women with induced abortion are at lower risk by virtue of the protection afforded by early childbirth, while those without induced abortion are at higher risk due to nulliparity or late childbirth. Consequently, adjustment for parity and age at first birth raises the relative risk estimate, as Ye et al observed. (The OR for the Ye et al cohort rises after adjustment for age and age at first birth from a raw value of 0.93 to 1.06) The opposite applies in a population such as in the US, where abortion is used predominantly to postpone first childbirth, rather than to limit family size. Another important difference, however, between these Chinese study populations and those of most western industrialised countries, is the very high prevalence of induced abortion in China. In the study of Ye et al, the prevalence of induced abortion is 51%, and in the study of Sanderson et al, it is 66%. The validity of any observed association – null or otherwise – between a given exposure and a given disease outcome, rests upon, among other things, the unexposed populations serving as a typical, appropriate reference group. Once the prevalence of a given exposure rises to a level of predominance, it is prudent to ask whether indeed the unexposed comparison group has instead become a subgroup, which is unexposed for some reason that bears relevance to its risk profile for the disease in question. In such a case, statistical adjustment cannot remove all such confounding, since the calculation of the adjustment term will necessarily be underestimated. In the case of the Shanghai study population, the confounding by parity and age at first birth would not be fully corrected for, and the relative risk for induced abortion would remain underestimated. Fortunately, the study design employed by Ye et al enables this hypothesis to be tested. In particular, the availability of a very large cohort of women (267 040) provided an ample supply of potential controls for the 702 eligible cases identified within the cohort. Indeed, Ye et al drew a control group that was closely age-matched by exact birth year. In both the cohort and case–control analyses, they found significant positive associations between age at first birth and breast cancer (data not shown) and nulliparity and breast cancer (relative risk (RR)=2.32, 95% CI: 1.45, 3.70 in the cohort analysis; case–control data not shown). As noted above, a nonsignificant association (RR=1.06) was found for induced abortion and breast cancer in both analyses. The hypothesis we propose in this letter, that the relative risk for induced abortion is underestimated in these analyses, can therefore easily be tested by drawing a new control group from the study cohort, wherein the controls are matched to cases not only by birth year, but also by parity and age at first birth. Such a case–control analysis, wherein controls are matched for these known confounders, would provide a more accurate estimate of the relative risk for induced abortion and breast cancer in this population. In addition, there is another line of evidence, touched upon by Ye et al, which supports our presently proposed hypothesis. In particular, Ye et al suggest that their observed prevalence of induced abortion in their cohort (i.e. 51%) may be an underestimate, as it is substantially lower than the 66% reported by Sanderson et al (2001) in their earlier study. However, this discrepancy can easily be explained by differences in timing. Sanderson et al, although their paper was published earlier, actually studied an overlapping, but younger cohort (than that studied by Ye et al) of Shanghai women. Specifically, Ye et al studied women born between 1925 and 1958, whereas Sanderson et al studied women born between 1932 and 1973. Since the ‘one-child policy’ is of relatively recent vintage, dating back only to 1980, it is to be expected that the prevalence of induced abortion would be substantially higher in the Sanderson et al study population. Assuming, then, that the observed prevalence of induced abortion in both the Ye et al and Sanderson et al studies are accurate, we would also expect that confounding due to the high induced abortion prevalence would be greater in the Sanderson et al study. Consequently, the magnitude of underestimation of the relative risk should also be greater, that is, the observed relative risk should be lower. This is in fact the case. Sanderson et al (2001) reported an odds ratio of 0.9 for parous women, and 1.0 for all women (Sanderson et al, 2000) in the two published reports of their study. Finally, the case made by Ye et al against there being a true positive association between induced abortion and breast cancer is not supported in the published record to the extent they suggest. They state: ‘No cohort studies (three are cited) or case–control studies nested within cohorts with ascertainment of abortion prior to development of breast cancer (two are cited) have shown associations of breast cancer with induced abortions.’ This claim is factually incorrect, since the prospective record-based case–control study of Howe et al (1989) – not cited at all by Ye et al – reported a statistically significant overall positive association (OR=1.9) between induced abortion and breast cancer. In fact, the overwhelming majority of published studies indicate a positive association between induced (but not spontaneous) abortion and breast cancer incidence (Brind et al, 1996). While it has been argued that some form of bias may be responsible for generating an apparent weak positive association (Lindefors-Harris et al, 1991), no credible evidence of such bias has been demonstrated. On the other hand, such confounding as we hypothesise in the present letter, can easily mask a true association, and we hope that Ye et al will take the opportunity to test for its presence in their analysis.

Direct radioimmunoassay of androstenediol-3-sulfate in the serum of normal men.

A commercially available antidihydrotestosterone antiserum was used for the direct radioimmunoassay of androstenediol-3-sulfate (ADS) in human serum. Aliquots of 1 or 2 microliter male serum (mean age of 40 subjects, 38.2 +/- 5.0 years) were diluted and extracted with ethanol for assay. The tracer, [7-3H]ADS, was prepared by sodium borohydride reduction of [7-3H]dehydroepiandrosterone sulfate (DS). Significantly cross-reacting steroids were testosterone, DS, androsterone sulfate, and epiandrosterone sulfate, which combined to produce a mean overestimation of ADS of 4.3 micrograms/dl in male serum. Mean serum ADS was 23.6 +/- 10.0 micrograms/dl (SD) in 20 fresh-frozen sera versus 28.4 +/- 9.7 micrograms/dl (SD) in 20 long-term (24.4 +/- 1.2 years) frozen specimens, showing stability on long-term frozen storage. Androstenediol-3-sulfate also showed a strong correlation with serum DS (r = 0.75). The possible physiologic significance of ADS is discussed, particularly in terms of the known estrogenicity of unconjugated androstenediol.

A new partition thin-layer chromatographic method for steroid separations

TLC plates with a 25 mu thick polyamide stationary phase were modified for the separation of neutral steroids by impregnation with propylene glycol. A mixture of tritiated 5 alpha-androstan-3 alpha,17 beta-diol, testosterone, 17 beta-hydroxy-5 alpha-androstan-3-one and 4-androstene-3,17-dione was applied to the plate and developed in a toluene mobile phase to a height of 13.6 cm. This resulted in complete resolution of the 4 compounds as detected by a gas flow scanner or imaging analyzer. Cutting and elution of peak areas with methanol resulted in quantitative recovery of all four steroids. The thinness of the layer also permitted a 3-5% counting efficiency on scanning, resulting in good quantitation of recovery without liquid scintillation counting. The high sorptive capacity of the polyamide layer also enabled extracts of normal human serum to be defatted on the TLC plate by development with pure hexane prior to the toluene step. The new method thus offers several advantages over existing methods for steroid separations and should be adaptable to separations of other relatively non-polar compounds.