No Reduced Recurrence Risk With Perioperative Aromatase Inhibitor Use in HR-Positive Breast Cancer

Ki67 Measurement Shows Prognostic Value

Get Permission

In the UK phase III POETIC trial reported in The Lancet Oncology, Ian Smith, MD, of The Royal Marsden NHS Foundation Trust, London, and colleagues, found that perioperative aromatase inhibitor therapy did not reduce the risk of recurrence in postmenopausal women with hormone receptor (HR)-positive early breast cancer vs no aromatase inhibitor therapy. The investigators also found that high tumor Ki67 values after 2 weeks of preoperative aromatase inhibitor therapy were associated with increased risk.1

Ian Smith, MD

Ian Smith, MD

Study Details

In the open-label multicenter trial, 4,480 postmenopausal women (modified intent-to-treat population) aged ≥ 50 years with HR-positive disease were randomly assigned 2:1 between October 2008 and April 2014 to receive perioperative aromatase inhibitor therapy consisting of letrozole at 2.5 mg/d or anastrozole at 1 mg/d for 14 days before and 14 days following surgery (n = 2,976) or no aromatase inhibitor therapy (control group, n = 1,504). Letrozole use was planned for 68% of patients and anastrozole for 30%. Adjuvant treatment was given according to UK standard local practice.

The primary endpoint was time to recurrence. Adjusted hazard ratios for efficacy outcomes were derived from multivariate models adjusted for progesterone receptor status, HER2 status, presurgical tumor grade, pathologic tumor size, presurgical histologic type, nodal status, age at randomization, and vascular invasion status.

A key second objective was investigation of the association between tumor Ki67 values and disease outcomes, with analysis conducted among patients with available paired Ki67 values. Ki67 was analyzed by immunohistochemistry in a core biopsy taken at baseline (Ki67B) and in either a core biopsy or the excision biopsy taken at surgery after 2 weeks of aromatase inhibitor treatment (Ki672W). Ki67 scores were considered low if < 10% and high if ≥ 10%.

For the aromatase inhibitor therapy vs control groups: 26% vs 31% received adjuvant chemotherapy, 76% vs 77% radiotherapy, and 7% vs 8% other adjuvant treatment (primarily anti-HER2 therapy); 88% of patients in both groups had HER2-negative disease and 11% vs 10% were HER2-positive; and 36% vs 34% underwent mastectomy and 64% vs 66%, conservative surgery.

Recurrence Risk

As of February 2018, median follow-up was 62.9 months (interquartile range = 58.1–74.1 months). Overall, breast cancer recurrence occurred in 280 patients in the aromatase inhibitor therapy group (9%) vs 154 (10%) in the control group (hazard ratio [HR] = 0.92, 95% confidence interval [CI] = 0.75–1.12, P = .40; adjusted HR = 0.96, 95% CI = 0.77–1.19, P = .70). The proportion free from breast cancer recurrence at 5 years was 91.0% vs 90.4%. Outcomes were consistent with the overall finding in subgroup analyses according to clinical characteristics, including nodal status.

No significant differences between the aromatase inhibitor therapy group and control group were observed for 5-year rates of relapse-free survival (87.9% vs 87.6%, HR = 0.94, P = .47; adjusted HR = 0.95, P = .59), local recurrence–free survival (98.6% vs 98.5%, HR = 0.86, P = .57; adjusted HR = 0.92, P =.75), distant recurrence–free survival (91.7% vs 90.9%, HR = 0.90, P = .30; adjusted HR = 0.94, P = .59), or overall survival (88.9% vs 88.9%, HR = 0.94, P = .50; adjusted HR = 0.91, P = .33). Second primary breast cancers occurred in 26 (< 1%) vs 24 (2%) of patients.

Analysis by Ki67 Status

A total of 2,528 patients in the aromatase inhibitor therapy group (678 in the control group) had paired Ki67 data available. After 2 weeks, significant reduction in Ki67 was observed in the aromatase inhibitor therapy group, with little change observed between paired samples in the control group. In the aromatase inhibitor therapy group, Ki672W was markedly lower among patients with HER2-negative tumors vs HER2-positive tumors (with a similar finding, with reduced magnitude of difference, in the control group).

Among aromatase inhibitor recipients with HER2-negative disease, the 5-year recurrence risk was 4.3% for those with low Ki67B and low Ki672W (n = 704), 8.4% for those with high Ki67B and low Ki672W (n = 1,097), and 21.5% for those with high Ki67B and high Ki672W (n = 406). Few patients had low Ki67B and high Ki672W (n = 28). The likelihood of recurrence was significantly higher in women with high Ki67B and high Ki672W vs those with high Ki67B and low Ki672W (HR = 2.59, P < .0001).

Among patients in the aromatase inhibitor group with HER2-positive disease, 5-year recurrence risk was 10.1% for patients with low Ki67B and low Ki672W (n = 32), 7.7% for those with high Ki67B and low Ki672W (n = 94), and 15.7% for those with high Ki67B and high Ki672W (n = 143). Few patients had low Ki67B and high Ki672W (n = 4). Likelihood of recurrence was numerically higher in women with high Ki67B and high Ki672W vs those with high Ki67B and low Ki672W (HR = 2.08, P = .093).


  • Aromatase inhibitor therapy for 14 days before and after surgery did not reduce risk of recurrence.
  • Patients receiving perioperative aromatase inhibitor therapy with high baseline and 2-week Ki67 were at greater risk of recurrence.

Adverse Events

The most commonly reported grade 3 adverse events (no grade 4 adverse events were observed) were musculoskeletal pain in 29 (1%) vs 13 (1%) of patients and hot flushes in 20 (1%) vs 6 (< 1%). Serious adverse events occurred in 11 patients in the aromatase inhibitor therapy group, with the most common being pulmonary embolism (n = 3) and musculoskeletal pain (n = 3). No treatment-related deaths were reported.

The investigators concluded: “[Perioperative aromatase inhibitor therapy] has not been shown to improve treatment outcomes but can be used without detriment to help select appropriate adjuvant therapy based on tumour Ki67. Most patients with low Ki67B or low [perioperative aromatase inhibitor therapy]-induced Ki672W do well with adjuvant standard endocrine therapy (giving consideration to clinical–pathological factors), whereas those whose [perioperative aromatase inhibitor therapy]-induced Ki672W remains high might benefit from further adjuvant treatment or trials of new therapies.” 

DISCLOSURE: The study was funded by Cancer Research UK. Dr. Smith has received honoraria from AstraZeneca, Eisai, Genomic Health, Pfizer, and Roche; has served as a consultant or advisor to AstraZeneca, Eisai, Genomic Health, Pfizer, and Roche; and has been reimbursed for travel, accommodations, or other expenses by Pfizer and Roche.


1. Smith I, Robertson J, Kilburn L, et al: Long-term outcome and prognostic value of Ki67 after perioperative endocrine therapy in postmenopausal women with hormone-sensitive early breast cancer (POETIC): An open-label, multicentre, parallel-group, randomised, phase III trial. Lancet Oncol 21:1443-1454, 2020.