Talcum Powder the Pluto’ of Prognostic Factors for Ovarian Cancer
By Robert L. Coleman, MD
Professor, University of Texas; M.D. Anderson Cancer Center, Houston
Dr. Coleman reports no financial relationships relevant to this field of study.
Synopsis: A l arge prospective cohort study of perineal talc use demonstrated no increased risk of ovarian cancer overall or within any histological subtype. In addition, no association with talc application method was observed.
Source: Houghton SC, et al. Pe rineal powder use and risk of ovarian cancer. J Natl Cancer Inst 2014;106:dju208 doi:10.1093/jnci/dju208.
Risk for ovarian cancer has been linked to talcum powder use for several years. Its structural properties and its historical link to asbestos have driven the biological plausibility. The preponderance of data to support this association has come from case-control studies; however, the only prospective cohort study from the Nurse’s Health Study did not show this effect, with the exception of serous invasive ovarian cancers. The current analysis from the Women’s Health Initiative Observational Study cohort prospectively assessed perineal powder use and risk of ovarian cancer. In this study, perineal powder use was assessed at baseline by self-report regarding application to genitals, sanitary napkins, or diaphragms and duration of use. The primary outcome was self-reported ovarian cancer centrally adjudicated by physicians. Cox proportional hazard regression was used to estimate risk, adjusting for covariates, including person-time until diagnosis of ovarian cancer (n = 429), death, loss to follow-up, or September 17, 2012. A total of 61,576 postmenopausal women without a history of cancer or bilateral oophorectomy were followed for a mean of 12.4 years. Fifty-three percent reported ever using perineal powder. Ever use of perineal powder (hazard ratio [HR] 1.06; 95% confidence interval [CI], 0.87-1.28) was not associated with risk of ovarian cancer compared with never use. Individually, ever use of powder on the genitals (HR = 1.12; 95% CI, 0.92-1.36), sanitary napkins (HR, 0.95; 95% CI, 0.76-1.20), or diaphragms (HR, 0.92; 95% CI, 0.68-1.23) was not associated with risk of ovarian cancer compared with never use, nor were there associations with increasing durations of use. There was no association between ever use and ovarian cancer histology, including invasive serous cancer. Estimates did not differ when stratified by age or tubal ligation status. Based on these findings, perineal powder use does not appear to influence ovarian cancer risk.
The discovery of prognostic factors in any disease serves not only to provide insight into disease pathogenesis, such as obesity and endometrial cancer with the implication of estrogen, but also to explore potential preventive interventions that can modulate risk, such as use of coagulation and cardiovascular disease with aspirin.1 In ovarian cancer, the most lethal of all the gynecologic cancers, prognostic factors carry added value as they highlight potential modifiable habits that might also impact mortality. As most are aware, ovarian cancer usually presents when the disease is widespread, causing symptoms of bloating, pelvic pressure, early satiety, and bladder dysfunction.2 Although these symptoms are frequently reported by patients in whom the diagnosis is ultimately made, the disconnection between a specific set of symptoms and stage of disease challenges any attempt to use this approach to modify mortality. Screening of otherwise normal women has also presented significant challenges for ovarian cancer. The disease is rare and has low prevalence even in menopausal women. This places substantial pressure on the performance of testing that would be utilized in a triage algorithm. The most frequently used screening modalities are the combination of examination, biomarkers such as CA125, and imaging such as transvaginal ultrasound. While these approaches have value in identifying women with the disease, the way in which they are implemented in an asymptomatic population, including recognition of abnormality (what’s abnormal?), frequency of testing (yearly? every 6 months? every 3 months?), and intervention of aberration (repeat assessment? referral? surgery?), is critical to the goal of identifying disease that is different from a non-screened population. In a disease like ovarian cancer, in which a clearly defined preinvasive state is not universally recognized or identifiable, the ultimate endpoint of a screening program is "stage shifting," or the alteration in the proportion of women diagnosed with earlier stage disease relative to the general population. Since stage I ovarian cancer is highly curable, this is a reasonable strategy to reduce mortality.
Unfortunately, stage I ovarian cancer is usually diagnosed by serendipity. Indeed, a report from the Prostate, Lung, Colorectal and Ovarian cancer screening trial demonstrated once-a-year screening with CA125 and transvaginal ultrasound not only did not increase the number of early-stage cases, but it increased morbidity due to complications from unnecessary surgery.3 The United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) randomized screening trial of more than 200,000 menopausal women assessing two different diagnostic triage algorithms (vs standard of care) in asymptomatic menopausal women has completed accrual and is expected to report in 2015. This trial’s primary endpoint is overall survival. An initial report of the prevalence data from the two screening algorithms demonstrated an efficiency and precision difference among women undergoing referral and surgical intervention.4 A separate prospective cohort study utilizing a two-step risk of ovarian cancer algorithm (ROCA), which, incidentally, is similar to one of the two screening strategies being utilized in the UKCTOCS trial, showed promise of the stage migration effect.5 In this trial, 4051 asymptomatic menopausal women underwent annual CA125 and utilized a ROCA mathematical algorithm to provide risk estimates of ovarian cancer. The resulting "low-," "intermediate-," and "high-risk" designation proscribed the next intervention, namely, repeat annual CA125, repeat CA125 in 3 months, and transvaginal ultrasound and gynecologic oncology referral, respectively. Ten women ultimately underwent surgical intervention, and four invasive ovarian cancers were found (40% positive predictive value; one stage IA, two stage IC and one stage IIC). While promising, proper evaluation of this approach will require the sample size, follow-up, and design (control group) of the UKCTOCS trial to assess the ultimate merit of screening in this disease.
Thus, in the absence of effective screening, attention has focused on prevention strategies. Many of these interventions, such as oral contraceptives, aspirin, salpingectomy, salpingo-oophorectomy, and tubal ligation, were identified as significant prognostic factors associated with reduced odds of diagnosis.1 As intuitive as these factors may seem and as easy as they are to identify, the business of properly assigning risk and the directionality of effect (positive, negative, or neutral) is much more difficult. In addition, the leap from identification of a prognostic factor to the effect of modulating risk by doing some sort of intervention (medication, surgery, habit alteration) based on that factor is a substantial gamble. Prognostic factors that accurately reflect the risk of developing a disease in a population require careful assessment of exposures. Most of the trials that serve to identify risk and the associated factors are done in retrospect and are subject to a profound effect of recall bias.6 It’s not hard to imagine that a woman with advanced stage ovarian cancer following surgery and chemotherapy might attribute blame to a specific habit, such as talcum powder use, and the amount of exposure differently relative to a woman without disease. Studies of oral contraceptive use, a noted prognostic factor associated with reduced risk of ovarian cancer, where centralized records of prescriptive practice exist, highlight this recall bias effect. In addition, accuracy of intended exposure, such as prescriptions made and actual use, provide another element of bias that is difficult to control. So while retrospective case-control trials are the primary resource from which prognostic factors are developed, they are often fraught with substantial bias hurdles that can lead to inconclusive or even disparate findings.
Such is the case with talcum powder. Talc is a water-absorbing mineral composed of magnesium silicate that has structural similarities and co-occurs with asbestos. The link of asbestos and cancer is relatively strong, so the implication of talc and cancer has been long suspected. The mechanism through which asbestos causes cancer is not completely understood, but its induction of a chronic inflammatory response and alteration in local immunogenicity to antigens in the microenvironment have been documented. Both of these factors have also been implicated in the carcinogenesis of ovarian cancer. Talc is a frequent component of genital powders and is usually applied directly on the perineal skin in a variety of ways. Historically, talcum powders used in cosmetics were not purified talc and had contamination with asbestos factors. However, in 1976, the Cosmetic, Toiletry and Fragrance Association (now known as the Personal Care Products Council) issued stringent purity standards for talc used in cosmetics, including specifications that talc must contain no detectable fibrous asbestos mineral. Nevertheless, talc fibers have been identified in the vagina, cervix, uterus, and ovaries in women who have reported perineal talc use. The quantity of these fibers is substantially reduced in ovarian tissue relative to the vagina and, while granulomatous inclusions have also been identified, the direct association of these foreign body reactions and cancer has not been observed. A comprehensive analysis of the safety assessment of talc used in cosmetics was conducted in 2006 by the International Agency for Research on Cancer’s Cosmetic Ingredient Review Expert Panel and released for public consumption 2010.7 In this report, toxicokinetics, preclinical and clinical toxicology, reproductive and developmental toxicity, genotoxicity, and carcinogenicity were extensively reviewed. Their concluding statement is summarized:
In 2010, the IARC Working Group determined that there is limited evidence in experimental animals for the carcinogenicity of talc not containing asbestos or asbestiform fibers . For humans, the evaluation of the IARC working group was that perineal use of talc-based body powder is possibly carcinogenic to humans (Group 2B), and that inhaled talc not containing asbestos or asbestiform fibers is not classifiable as to its carcinogenicity (Group 3). In evaluating the carcinogenicity of talc in humans, the Working Group reviewed cohort studies of talc miners and millers, cohort and case-controlled studies examining the association of cosmetic talc use and the risk of ovarian cancer in humans, and the animal data and evidence regarding the potential mechanisms through which talc might cause cancer in humans. The Working Group found there is inadequate evidence in humans for the carcinogenicity of inhaled talc not containing asbestos or asbestiform fibers and there is limited evidence in humans for the carcinogenicity of perineal use of talc-based body powder.
The strength of their Group 2B conclusion rested on the volumes of retrospective reports, including a meta-analysis of 20 case-control studies and a pooled analysis of eight other population-based, case-control studies implicating a risk of up to 35% between talc use in perineal powders and ovarian cancer. However, the only two prospective cohort studies, including the current trial from the Women’s Health Initiative Observational Study, provided no association.8 The current trial is the largest prospective trial to assess the implied risk and is strengthened by it low risk of recall bias. However, only data on duration of use (vs duration and frequency of use) were available. To date, a dose-response relationship has not been made.
In again out again what conclusions/recommendations can be made about talc and ovarian cancer? One clear assurance is that talcum powder used in cosmetics is regulated to be asbestos free. Second, evidence of migration of talc fibers from the perineum to the fallopian tubes and ovaries is present, but is devoid of the asbestos-inducing inflammatory response, disrupting the biological plausibility of talc exposure and cancer. Third, the strength of association, if present at all, is weak and the current study’s design and conclusions should be reassuring to users. Finally, it is unlikely that modifying exposure to this "Pluto of a prognostic factor" will modulate any potential diagnostic risk or mortality from ovarian cancer.
- Coleman RL, et al. Nat Rev Clin Oncol 2013;10:211-224.
- Goff BA, et al. Cancer 2007;109:221-227.
- Buys SS, et al. JAMA 2011;305:2295-2303.
- Menon U, et al. Lancet Oncol 2009;10:327-340.
- Lu KH, et al. Cancer 2013;119:3454-3461.
- Schulz KF, Grimes DA. Lancet 2002;359:431-434.
- IARC Working Group on the Evaluation of Carcinogenic Risks to Humans. Available at: http://monographs.iarc.fr/ENG/Monographs/vol93/mono93.pdf. Accessed Nov. 1, 2014.
- Gertig DM, et al. J Natl Cancer Inst 2000;92:249-252.