New Studies Explore Exposure to Cancer-causing Agents

Get Permission

Key studies on cancer epidemiology and prevention delivered both reassuring and not-so-reassuring findings on exposure to agents believed to be cancer-promoting. Kala Visvanathan, MD, MHS, of The Sidney Kimmel Cancer Center, Johns Hopkins School of Medicine and School of Public Health, Baltimore, put the abstracts into perspective at the Best of ASCO Boston meeting.

Combined Hormone Replacement Increases Breast Cancer Risk

More data presented at ASCO confirmed previous findings that combined estrogen-plus-progestin use increases breast cancer risk, while another study concluded that this risk cannot be attenuated by low-dose tamoxifen.

An observational study included a cohort of 41,449 postmenopausal women gleaned from the Women’s Health Initiative who used estrogen plus progestin (N = 16,121) or no hormone replacement therapy (N = 25,328).1 The average participant had received hormone replacement therapy for 5.3 years prior to cohort entry.

Multivariate analysis showed invasive breast cancer incidence to be 55% higher in the estrogen-plus-progestin users (P < .001) after taking into consideration potential confounders. Survival following breast cancer was similar between hormone replacement therapy users and nonusers, but breast cancer mortality was increased in the hormone replacement group (P < .001).

‘Gap Time’

The findings from the Women’s Health Initiative observational study were consistent with other observational studies and the clinical trial in showing that combined hormone replacement therapy is associated with an increased risk of cancer—approximately 1.5- to 2.5-fold over 10 years. It is unclear when the risk actually begins to rise, whether upon initiation of hormone replacement therapy or several years later.

The concept of “gap time” was substantiated in the observational study, which showed that the risk is higher if hormonal therapy is initiated close to the onset of menopause (HR = 1.45), but is attenuated if hormonal therapy is begun 5 years or more after menopause (HR = 1.19). The gap time effect is persisted in analyses adjusted for duration of estrogen-plus-progestin use, showing that initiation of the combination 15 years postmenopause still carried a risk higher than was seen with nonusers.

“Further studies are needed to determine how we might exploit this gap time concept,” she said.

Dr. Visvanathan reminded the audience that they need to be aware of the patient characteristics when interpreting data from the Women’s Health Initiative hormone trials evaluating hormone replacement therapy. The average age at entry was 63, indicating that women were initiated on hormone replacement therapy much later after the onset of menopause than is typically seen outside of a trial. Also, 80% were at least 10 years beyond menopause, 26% had previously taken hormone replacement therapy prior to enrolling, and treatment was discontinued early in 40% in the combined-hormone arm and in 54% of the estrogen-only arm in the initial trial.

Low-dose Tamoxifen to Oppose Hormone Therapy

The phase III Italian HOT trial was a primary breast cancer prevention trial that aimed to determine if low-dose tamoxifen (5 mg/d for 5 years, vs placebo) might mitigate the risk for breast cancer posed by hormone replacement therapy.2 The rationale was backed by data from collaborative groups, tamoxifen prevention trials, and neoadjuvant studies.

“The concept was interesting. The aim was to identify an approach whereby women could have the benefits of hormone replacement therapy while minimizing breast cancer risk,” Dr. Visvanathan said.

Unfortunately, accrual coincided with the announcement from the Women’s Health Initiative that implicated hormone replacement therapy in the development of breast cancer. Therefore, target enrollment was downsized from 8,500 to 4,500 subjects.

The study found that the combined use of low-dose tamoxifen and hormone replacement therapy may retain the benefit of each drug while reducing breast cancer risk. However, the 20% reduction in risk was not statistically significant. A subanalysis showed greater reduction (51%) in the cohort treated for at least 12 months, though again, this was not statistically significant. Nevertheless, significant effects were seen in protection against luminal A cancers (a 67% reduction, P = .019) and progesterone-positive tumors (a 57% reduction, P = .038).

Adverse effects were those well known for tamoxifen: endometrial polyps, hot flashes, night sweats, vaginal discharge, and dryness.

“The data overall were not statistically significant, but there were suggestive differences by subtype,” Dr. Visvanathan said. “This is an interesting result, but the size of the study was reduced. I look at this study more as promising but not definitive, and definitely not practice-changing.”

There is a need, she added, for studies to identify subgroups of patients who might benefit from and safely receive hormone replacement therapy, and also to determine whether such treatment can be safely delivered to breast cancer survivors when required.

Dexrazoxane Does Not Increase Cancer Risk

Reassuringly, investigators showed that the dexrazoxane, a cardioprotectant, is not associated with an increased risk of secondary acute myeloid leukemia (AML). The potential risk of secondary AML in children exposed to dexrazoxane during cancer treatment has limited its use in the United States. The potential for risk was evaluated in a retrospective cohort study of 15,532 children exposed to anthracyclines, with and without dexrazoxane.3 The incidence of secondary AML was only 0.52%, and was no different between those exposed to dexrazoxane or not. Mortality was also no different.

The strength of the study is its use of a large administrative database, which enabled investigators to examine various settings and patterns of drug use. The diagnosis of AML in only 80 individuals indicates how rare the condition is, and how important it is to have access to well annotated large databases, she said.

Thiazolidinediones Do Not Increase Bladder Cancer

The FDA has warned that the use of pioglitazone, a thiazolidinedione used to treat type 2 diabetes, may increase the risk of bladder cancer, but information about these drugs as a class has been lacking. Philadelphia investigators conducted a cohort study of patients with type 2 diabetes who initiated treatment with a thiazolidinedione (n = 18,459) or sulfonylureas (n = 41,396), a similar drug class, in The Health Improvement Network, a general practitioner database in the United Kingdom.4 After 2 years’ median duration of treatment, patients receiving thiazolidinediones had no greater incidence of bladder cancer than those on sulfonylureas.

Subanalyses did show that ≥ 5 years of exposure to thiazolidinediones may be associated with an increased risk of bladder cancer, but there was no dose response and this is not consistent with prior studies and therefore should be interpreted with caution, she said.

Dr. Visvanathan commented that the size of the study, and the fact that there was a reasonable comparator, gives validity to the results. Lack of details on dose and other confounders were definite limitations.

In closing, she pointed out that observational studies and randomized trials both have a place in improving clinical practice. As is the case for clinical trials, clinicians need to be able to differentiate between those observational studies that are well done and those that are not, she cautioned. ■

Disclosure: Dr. Visvanathan reported no potential conflicts of interest.


1. Chlebowski RT, Anderson GL, Kuller LH, et al: Estrogen plus progestin and breast cancer incidence and mortality. 2012 ASCO Annual Meeting. Abstract 1501. Presented June 3, 2012.

2. Bonanni B, Maisonneuve P, Serrano D, et al: A phase III prevention trial of low-dose tamoxifen in HRT users: The HOT trial. 2012 ASCO Annual Meeting. Abstract 1500. Presented June 3, 2012.

3. Walker DM, Li Y, Huang Y-S, et al: Dexrazoxane exposure and risk of secondary acute myeloid leukemia in pediatric cancer patients. 2012 ASCO Annual Meeting. Abstract 1504. Presented June 3, 2012.

4. Mamtani R, Haynes K, Bilker WB, et al: Long-term therapy with thiazolidinediones and risk of bladder cancer: A cohort study. 2012 ASCO Annual Meeting. Abstract 1503. Presented June 3, 2012.