NSABP P-1

Published

Fisher B, et al. "Tamoxifen for Prevention of Breast Cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study". Journal of the National Cancer Institute. 1998. 90(18):1371-1388.
PubMed • Full text • PDF

Clinical Question

Does tamoxifen reduce the incidence of breast cancer among women with an increased risk for breast cancer?

Bottom Line

Among women with an increased risk for breast cancer, tamoxifen reduces the risk of ER-positive breast cancer, but it provides no mortality benefit. Tamoxifen increases the risk of endometrial cancer and pulmonary embolism.

Major Points

The selective estrogen receptor modulator (SERM) tamoxifen enhances the effects of estrogen on the uterus, bone, and coagulation system, but selectively blocks the effects of estrogen in the breast. It was first FDA-approved for use in women with breast cancer in 1977, and its effects in women with hormone receptor-positive breast cancer were first demonstrated in the 1989 NSABP B-14 trial. Despite this long track record in breast cancer treatment, its role in the primary prevention of breast cancer had not yet been studied.

The National Surgical Adjuvant Breast and Bowel Project, Prevention-1 (NSABP P-1) study, also called the Breast Cancer Prevention Trial (BCPT), was the first randomized trial to demonstrate tamoxifen's effectiveness in the primary prevention of breast cancer among women at higher risk. NSABP P-1 studied 13,388 women with three risk factor profiles: 1) age ≥60 years, 2) age 35-59 years with a Gail model 5-year predicted risk for breast cancer ≥1.66%, or 3) age 35-59 years with a history of LCIS. Participants were randomized to tamoxifen or placebo. At a mean follow-up of 4 years, tamoxifen was associated with 49% fewer invasive breast cancers and 69% fewer ER-positive breast cancers among premenopausal and postmenopausal women. Tamoxifen did not impact overall survival, and was associated with 3-fold increased risk of pulmonary embolism (1 per 1,000 patient-years) and a 4-fold increased risk of endometrial cancer among postmenopausal women (3 per 1,000 patient-years).

Despite these adverse events, the results of NSABP P-1 were well received. In 1998, the FDA requested early NSABP P-1 data to expedite the FDA approval of tamoxifen for primary breast cancer prevention, and in 1999, the agency approved tamoxifen for breast cancer prevention. The follow-up NSABP P-2 STAR trial (2006) compared raloxifene to tamoxifen, and demonstrated similar effectiveness of raloxifene among postmenopausal women, with fewer episodes of endometrial cancer, pulmonary embolism, and cataracts compared to tamoxifen.

NSABP P-1 provides the only data on breast cancer chemoprevention among BRCA1 or BRCA2 mutation carriers. A separately published analysis revealed that of the 288 evaluable NSABP P-1 participants who developed breast cancer, 19 (6.6%) had an associated mutation in BRCA1 (n=8) or BRCA2 (n=11). The authors estimate that tamoxifen decreased the risk of breast cancer among BRCA2 mutation carriers (RR 0.38; 95% CI 0.06-1.56) but not among BRCA1 mutation carriers (RR 1.67; 95% CI 0.32-10.70). The subgroup analysis was retrospective and its numbers are very small, but this analysis remains the best evidence of a chemopreventive benefit of tamoxifen for BRCA mutation carriers.

Guidelines

ASCO Breast Cancer Risk Reduction Guidelines (2013)

• Tamoxifen should be discussed as an option to reduce the risk of invasive breast cancer, specifically ER-positive breast cancer, in pre- and postmenopausal women who are age ≥35 years with a 5-year projected absolute breast cancer risk ≥1.66% or with LCIS. Risk reduction benefit continues for at least 10 years.

Design

• Multicenter, double-blind, parallel group, randomized, placebo-controlled trial
• N=13,388 women at increased risk of developing breast cancer
  • Tamoxifen (n=6,681)
  • Placebo (n=6,707)
• Setting: 131 centers in the US and Canada
• Enrollment: 1992-1997
• Mean follow-up: 4 years
• Analysis: Intention-to-treat

Population

Inclusion Criteria

• One of
  • Age ≥60 years
  • Age 35-59 with a Gail model 5-year risk of breast cancer ≥1.66%
  • Age 35-59 with a history of LCIS
• Life expectancy ≥10 years
• No evidence of breast cancer by clinical breast exam or mammogram within prior 180 days
• Endometrial sampling prior to randomization if uterus present and randomized after July 1994

Exclusion Criteria

• Estrogen or progesterone replacement therapy, OCPs, or androgens within prior 3 months
• History of DVT or PE
• Abnormal WBC, platelet count, hepatic function, renal function tests
• Pregnancy or plans to become pregnant while on study drug

Baseline Characteristics

• Age in years:
  • 35-39: 2.6%
  • 40-49: 36.7%
  • 50-59: 30.8%
  • 60-69: 24%
  • ≥70: 6%
• White: 96.4%
• First-degree relatives with breast cancer:
  • 0: 23.8%
  • 1: 56.8%
  • 2: 16.4%
  • ≥3: 3%
• Prior hysterectomy: 37.1%
• History of LCIS: 6.3%
• History of atypical hyperplasia: 9%
• Gail model 5-year predicted risk:
  • ≤2%: 25%
  • 2-3%: 31%
  • 3-5%: 26.6%
  • ≥5%: 17.4%

Interventions

• Breast cancer risk assessments via Gail Model
• Randomized to tamoxifen 20mg or placebo daily for 5 years

Outcomes

Comparisons are tamoxifen vs. placebo, presented as annual rates per 1,000 women.

Primary Outcomes

Invasive breast cancer

3.43 vs. 6.76 (RR 0.51; 95% CI 0.39-0.66)

Secondary Outcomes

ER-positive breast cancer

1.58 vs. 5.02 (RR 0.31; 95% CI 0.22-0.45)

ER-negative breast cancer

1.46 vs. 1.20 (RR 1.22; 95% CI 0.74-2.03)

Non-invasive breast cancer

1.35 vs. 2.68 (RR 0.50; 95% CI 0.33-0.77)

Mortality

2.17 vs. 2.71 (RR 0.81; 95% CI 0.56-1.16)

Adverse Events

Invasive endometrial cancer in women ≥50 yrs

3.05 vs. 0.76 (RR 4.01; 95% CI 1.70-10.90)

MI

1.19 vs. 1.07 (RR 1.11; 95% CI 0.65-1.92)

Bone fractures

4.29 vs. 5.28 (RR 0.81; 95% CI 0.63-1.05)

Stroke

1.45 vs. 0.92 (RR 1.59; 95% CI 0.93-2.77)

PE in women ≥50 yrs

1.00 vs. 0.31 (RR 3.19; 95% CI 1.12-11.15)

DVT

1.34 vs. 0.84 (RR1.60; 95% CI 0.91-2.86)

Cataracts

24.82 vs. 21.72 (RR 1.14; 95% CI 1.01-1.29)

Subgroup analysis

Primary endpoint remains significant for age and history of atypical hyperplasia.

History of LCIS

No: 3.30 vs. 6.41 (RR 0.51; 95% CI 0.39-0.68)

Yes: 5.69 vs. 12.99 (RR 0.44; 95% CI 0.16-1.06)

Five year predicted breast cancer risk

≤2%: 2.06 vs. 5.54 (RR 0.37; 95% CI 0.18-0.72)

2-3%: 3.51 vs. 5.18 (RR 0.68; 95% CI 0.41-1.11)

3-5%: 3.88 vs. 5.88 (RR 0.66; 95% CI 0.39-1.09)

≥5%: 4.52 vs. 13.28 (RR 0.34; 95% CI 0.19-0.58)

First degree relatives with breast cancer

0: 2.97 vs. 6.45 (RR 0.46; 95% CI 0.24-0.84)

1: 3.03 vs. 6.00 (RR 0.51; 95% CI 0.35-0.73)

2: 4.75 vs. 8.68 (RR 0.55; 95% CI 0.30-0.97)

≥3: 7.02 vs. 13.72 (RR 0.51; 95% CI 0.15-1.55)

Funding

• Funded by the National Cancer Institute, National Institutes of Health, and Department of Health and Human Services.
• Study drugs supplied by Zeneca Pharmaceuticals