February 29, 2008 — Combination hormone therapy may compromise the diagnostic performance of mammograms and breast biopsies, according to a study published in the February 25 issue of the Archives of Internal Medicine. Even after hormone therapy was discontinued, the higher frequency of mammograms with abnormal findings continued for at least 12 months.

"Clinicians should be aware that breast cancer diagnosis is hindered by combined hormone therapy and the problem persists for at least a year after stopping its use," said lead author Rowan T. Chlebowski, MD, PhD, an investigator at the Los Angeles Biomedical Research Institute at Harbor–UCLA Medical Center, Torrance, California. "The risk of an abnormal mammogram was significant up to 12 months, but problems still didn't completely clear until 2 years after hormone therapy was halted."

The Women's Health Initiative (WHI) previously reported that combination conjugated equine estrogens plus medroxyprogesterone acetate was associated with an increased risk for invasive breast cancers in menopausal women. Results from this large, randomized trial also showed that combined hormonal therapy significantly increased the number of abnormal mammograms and breast cancers that were larger and diagnosed at a more advanced stage.

"We have already found that combined hormone therapy increases the incidence of breast cancer, and that these cancers are likely to be in a more advanced stage," Dr. Chlebowski said in an interview. "We started to think about mammograms, and if diagnosis was going to be hindered by hormone therapy. We had an idea that it might be a problem."

Observational studies that evaluated the effect of combined hormone therapy on the detection of breast cancer have yielded mixed results. In the current study, Dr. Chlebowski and colleagues examined the impact of combination hormone therapy on breast cancer detection in a cohort of 16,608 postmenopausal women who participated in the WHI clinical trial, beginning in 1993 through 1998.

The participants were randomized to either a treatment group (combination of equine estrogens at 0.625 mg a day and medroxyprogesterone acetate at 2.5 mg a day) or a placebo group. The women underwent a baseline mammogram and breast examination and then annually during the course of the study period. Biopsies were performed based on clinical findings. The researchers evaluated the effects of combined hormone therapy for 5.6 years, using receiver operating characteristic (ROC) curve analyses to evaluate mammography results.

They found a significantly higher frequency of mammograms with abnormal results among women using combination hormone therapy vs placebo (35.0% vs 23.0%, respectively; P < .001). Women using hormone therapy also had an approximately 4% greater risk of having an abnormal mammogram result after 1 year, and that number increased to approximately 11% after 5 years.

Although the specificity and the negative predictive value of mammograms were only slightly lower among women using hormone therapy, the sensitivity and the positive predictive values were compromised by their use.

The number of breast cancers was significantly increased and diagnosed at more advanced stages in the hormone group, but biopsies that were performed in these women demonstrated breast cancer less frequently. A total of 1220 biopsies were performed in women using combined hormone therapy, of which 180 (14.8%) results yielded a diagnosis of breast cancer. In the placebo group, 672 biopsies were performed and breast cancer was diagnosed in 132 (19.6%).

"The women using hormones had more cancers, but they were missed more often than in the placebo group," Dr. Chlebowski told Medscape Oncology. "This is a therapy that not only increases the risk of cancer, but makes it harder to find it."

Hormone therapy also increases breast density, although the biologic significance of these changes or their effect on mammographic interpretation has not been established. Some observational studies have suggested that a higher breast density is associated with delays in diagnosis. Breast density is not the only significant factor, Dr. Chlebowski explained and, rather, is just 1 part of the puzzle.

Among women aged 50 to 59 years, the increase in mammograms with abnormal results among those using hormones was the same as for the overall group. Compared with the placebo group, their time to first biopsy was significantly shorter, and the percentage undergoing a biopsy after 5 years was significantly greater. They also had a 3% higher risk for abnormal mammographic results after 1 year and a 9% greater risk after 5 years.

"Women considering combined hormone therapy should be aware that they will have about a 1 in 10 and 1 in 25 chance of having an otherwise avoidable mammogram or breast biopsy, respectively," said Dr. Chlebowski.

Although other imaging techniques are available, such as digital mammography (more accurate than standard mammography in women with dense breast tissue), magnetic resonance, or ultrasonography, it is yet unclear if they are more effective than mammography in women using combination hormone therapy.

The WHI program is funded by the National Heart, Lung, and Blood Institute, US Department of Health and Human Services. Dr. Chlebowski is a consultant for AstraZeneca, Novartis, Organon, and Eli Lilly and Co and has previously received grant support from Eli Lilly. Co-author Robert D. Langer, MD, MPH, is an expert witness for Wyeth Pharmaceuticals. Two other study authors have obtained funding.

Arch Intern Med. 2008;168:370-377.

Clinical Context

Study Highlights

• The WHI enrolled 16,608 postmenopausal women at 40 clinical centers for 5 years.
• Eligibility criteria were ages 50 to 79 years, no history of hysterectomy, and baseline mammogram and breast examination results not suggestive of cancer.
• Women taking postmenopausal hormones at entry were eligible after a 3-month washout period.
• Exclusion criteria were history of breast cancer, other cancer within 10 years (except nonmelanoma skin cancer), and medical conditions likely to be fatal within 3 years.
• 8506 women were randomized to receive a daily single tablet of 0.625 mg of conjugated equine estrogens and 2.5 mg of medroxyprogesterone acetate.
• 8102 women received placebo.
• Median age was 63 years.
• Hormone and placebo groups were similar in age, ethnicity (84% white), educational level, age at menarche, body mass index, physical activity, alcohol intake, smoking, Gail risk score, number of term pregnancies, age at first birth, number of breast-fed children, number of first-degree relatives with breast cancer, benign breast disease and biopsies, and menopausal hormone therapy use.
• Follow-up included contact at 6 weeks to evaluate symptoms and adherence, interview every 6 months to evaluate clinical outcomes, and annual clinic visits for breast examination and mammograms.
• Self-reports of breast cancer were confirmed by medical records and pathology reports.
• After termination of study medications, follow-up continued every 6 months, and mammogram data were recorded for median of 2.4 years.
• True-positive mammogram result was defined by diagnosis of invasive breast cancer or ductal carcinoma in situ within 12 months of mammogram.
• True-positive breast biopsy result was defined by diagnosis of invasive breast cancer in the 6-month interval of biopsy report or within 2 months after interval when biopsy result was reported.
• Mammograms with abnormal results occurred more in hormone vs placebo group (35% vs 23%; P < .001).
• Risk of having mammogram with abnormal results was greater in hormone group by 4% after 1 year and by 11% after 5 years (P < .001).
• Mammogram sensitivity and positive predictive value were reduced in hormone vs placebo group.
• ROC curve and area under the curve (AUC) assessed mammogram diagnostic performance by study period and by study group, with AUC of 0.5 indicating nonpredictive model vs AUC of 1.00 indicating predictive model:
• • Mammograms were worse in identifying cancer in hormone vs placebo group for years 1 and 2 (ROC AUC, 0.85 vs 0.90; P = .02 for abnormal vs normal mammogram results in hormone vs placebo group) and years 3 and 4 (ROC AUC, 0.84 vs 0.94;
    P = .001).
  • For mammograms performed after 5 years, ROC AUC was 0.92 vs 0.94 (P = .15) for abnormal vs normal mammogram results in hormone vs placebo group.
• Mean (SD) time to first biopsy was shorter in hormone vs placebo group (3.4 [1.89] years vs 3.6 [1.8] years; P < .001).
• Cumulative percentage of women with biopsy was higher in hormone vs placebo group (10.0% vs 6.1%; P < .001).
• Biopsies diagnosed breast cancer less frequently in hormone vs placebo group
  (180 [14.8%] of 1220 biopsies vs 132 [19.6%] of 672 biopsies; P = .006).
• Younger women aged 50 to 59 years had similar findings as overall group.
• After study medication was discontinued, mammograms with abnormal results were more common in hormone vs placebo group at up to 6 months (P < .001) and 7 to 12 months
  (P = .005) but not significantly different for the 2 groups at 13 to 18 months and 19 to 24 months.

Pearls for Practice

• Postmenopausal women who use combined hormone therapy vs placebo have increased frequency of mammograms with abnormal results, with less sensitivity and positive predictive value and increased frequency of breast biopsy, which demonstrates cancer less frequently.
• Postmenopausal women who use combined hormone therapy vs placebo have an increased frequency of mammograms with abnormal results up to 12 months after the discontinuation of hormone therapy.

CME/CE Test