Clinical Validation

The first clinical validation of the Oncotype DX® Breast Recurrence Score for patients with invasive, node-negative breast cancer was prospectively designed study performed on a cohort of 668 patients who had participated in the NSABP B-14 study.1 The primary objective of this study was to determine whether the Breast Recurrence Score result was predictive of 10-year distant recurrence. Among the 668 tamoxifen-treated patients with ER+, node-negative disease, 51%, 22%, and 27% fell into the low-, intermediate-, and high- risk Recurrence Score groups, respectively. Analyses demonstrated that the Breast Recurrence Score result was a predictor of distant recurrence, independent of age and tumor size. The 10-year rate of distant recurrence was significantly lower in the low-risk group (6.8%) compared with the high-risk group (30.5%, p<0.001; Table 1). The Breast Recurrence Score result was also significantly predictive of relapse-free interval and overall survival (p<0.001 for both). In a multivariate Cox proportional hazards analysis, the Breast Recurrence Score result and tumor grade were the only significant predictors of distant recurrence (p<0.001).

Table 1. NSABP B-14 Clinical Validation Study in Node-Negative Patients (N=668):
Rate of Distant Recurrence at 10 Years by Risk Group
Risk Group
(Recurrence Score)
% of Patients
(N)
10-Year Rate of
Distant Recurrence
95% Confidence
Interval
Low (<18) 51% (338) 6.8% 4.0%, 9.6%
Intermediate (18-30) 22% (149) 14.3% 8.3%, 20.3%
High (=31) 27% (181) 30.5% 23.6%, 37.4%

A subsequent prospective study evaluated whether the Breast Recurrence Score result can predict late (5-15 years) distant recurrence risk in patients with high ER expressing tumors.2 The study cohort included 497 patients enrolled in the original B-14 trial with quantitative ER expression above 9.1. The Breast Recurrence Score risk groups showed a significant association with late distant recurrence (log rank p=0.01). Patients with low Breast Recurrence Score results had a recurrence risk of 4.7% at 5–10 years and 6.8% at 5–15 years. While those with high Breast Recurrence Score results had a recurrence risk of 12.6% at 5-10 years and 16.4% at 5-15 years. The authors concluded that these results suggest that extending endocrine therapy beyond 5 years may be most beneficial in those patients with high (and intermediate) Breast Recurrence Score results with higher quantitative ER expression.

A prospectively-designed clinical validation study of the Oncotype DX Breast Recurrence Score was performed on a cohort of 651 patients with ER+, node-negative breast cancer from the NSABP B-20 trial.3 The objective of this study was to determine whether the Breast Recurrence Score predicted the magnitude of chemotherapy benefit. In this trial, tamoxifen-treated patients had been randomized to receive either no chemotherapy (n=227) or chemotherapy (cyclophosphamide, methotrexate, and 5-fluorouracil [CMF] or methotrexate and 5-fluorouracil [MF]; n=424). Upon analysis of clinical variables, including age, tumor size, grade (assigned by three independent pathologists), and hormone receptor status, the Breast Recurrence Score value was the strongest predictive factor of chemotherapy benefit and was the only variable that had a statistically significant interaction with chemotherapy treatment (p=0.038).

The results of this study demonstrated that the Breast Recurrence Score result was predictive of the magnitude of chemotherapy benefit. Patients with low scores had minimal, if any benefit from chemotherapy, compared to patients with high scores, who demonstrated a significant benefit. The group of 218 patients with low Breast Recurrence Score results treated with tamoxifen and chemotherapy showed minimal if any chemotherapy benefit when compared with 135 patients with low-risk results treated with tamoxifen only (p=0.61; Table 2). In contrast, among patients with high Recurrence Score results, there was a significant treatment benefit in the 117 patients who received chemotherapy compared to 47 patients treated with tamoxifen alone (p<0.001).

Table 2. NSABP B-20 Clinical Validation Study in Node-Negative Patients (N=651):
Proportion of Patients Free of Distant Recurrence at 10 Years by Risk Group
Recurrence Score
Risk Group
% of Patients
(N)
Distant Recurrence Free at
10 Years (%)
95% Confidence
Interval
Tamoxifen Alone
All Patients 227 87.8% 83.3%, 92.3%
Low (<18) 59% (135) 96.8% 93.7%, 99.9%
Intermediate (18-30) 20% (45) 90.9% 82.5%, 99.4%
High (=31) 21% (47) 60.5% 46.2%, 74.8%
Tamoxifen Plus Chemotherapy
All Patients 424 92.2% 89.4%, 94.9%
Low (<18) 51% (218) 95.6% 92.7%, 98.6%
Intermediate (18-30) 21% (89) 89.1% 82.4%, 95.9%
High (=31) 28% (117) 88.1% 82.0%, 94.2%

A further validation study was conducted in samples from 1,231 post-menopausal patients with ER+ disease enrolled in the ATAC trial.4 Of these patients, 872 had node-negative and 306 had node-positive disease. The Breast Recurrence Score result was shown to be a statistically significant predictor of distant recurrence in patients with node-negative (p<0.001) or node-positive (p=0.002) disease treated with either tamoxifen or anastrozole. In a multivariate analysis of node-negative patients that included tumor size, central grade, and age, the Breast Recurrence Score result and tumor size were the only variables that were statistically significant in predicting time to distant recurrence (p<0.001). When adjusted for other clinical variables, the Breast Recurrence Score result was significantly predictive for overall survival, with a hazard ratio of 2.5 (95% CI: 1.5 to 4.0) for the high- vs low-risk Breast Recurrence Score groups.

The performance of the Oncotype DX assay in predicting breast cancer-specific mortality in patients with node-negative disease has been reported by Habel, et al.5 The study was performed within a community hospital–based patient population from multiple sites within the Kaiser Permanente system, and included 220 patients who had died of breast cancer and 570 matched controls. After adjusting for tumor size and grade, the Breast Recurrence Score result was significantly associated with risk of breast cancer-specific death in ER+, tamoxifen-treated and -untreated patients (p=0.003 and p=0.03, respectively). At 10 years, the risks for breast cancer death in tamoxifen-treated patients with ER+ disease (55 cases and 150 controls) were 2.8%, 10.7%, and 15.5% for those in the low-, intermediate-, and high-risk Breast Recurrence Score groups, respectively. Among patients with ER+ disease who were not treated with tamoxifen (110 cases and 251 controls), the 10-year risks for breast cancer death were 6.2%, 17.8%, and 19.9% for those in the low-, intermediate- and high-risk Breast Recurrence Score groups, respectively.

Toi et al. conducted a validation study in 200 tamoxifen-treated ER+, node-node negative patients enrolled in a Japan Breast Cancer Translational Research Group trial.6 Patients in the low-risk Breast Recurrence Score group had a significantly lower risk of distant recurrence than patients in the high-risk Breast Recurrence Score group (p<0.001), with 10-year rates of 3.3% and 24.8%, respectively. In a Cox proportional hazards model of distant recurrence-free interval, the continuous Breast Recurrence Score result remained significantly associated with the risk of distant recurrence (p<0.001) when the model was adjusted for age (<50 vs ≥50 years) and tumor size (≤2 vs >2 cm). In addition, patients in the low-risk Breast Recurrence Score group had significantly longer overall survival (p=0.008) compared to patients in the high-risk Breast Recurrence Score group.

Additional evidence confirming the assays validation is from analyses of tumor blocks collected in the course of the ECOG 2197 trial. Analyses were performed in a total of 465 patients with hormone receptor-positive disease, including both pre- and post-menopausal patients and both node-negative and node-positive (1-3 positive nodes) disease, all treated with anthracycline-based chemotherapy (AC or AT).7 Patients with a low Breast Recurrence Score result had lower five-year recurrence rates than patients with high scores (node-negative 4% vs 13%; node-positive 5% vs 25%). This finding was consistent among the node-negative and node-positive patient subsets. At ten -year follow up, the Breast Recurrence Score continued to be a highly significant predictor of distant recurrence (p<0.0001).

 

REFERENCES

1. Paik S, Shak S, Tang G, et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med. 2004; 351 (27): 2817-26.
2. Wolmark N, Mamounas EP, Baehner FL, et al. Recurrence Score and Quantitative ER Expression Predicts Late Distant Recurrence Risk in Estrogen Receptor Positive Breast Cancer after Five Years of Tamoxifen. Presented at American Society of Clinical Oncology Annual Meeting; May 2014; Chicago, IL.
3. Paik S, Tang G, Shak S, et al. Gene expression and benefit of chemotherapy in women with node-negative, estrogen receptor-positive breast cancer. J Clin Oncol. 2006; 24 (23): 3726-34.
4. Dowsett M, Cuzick J, et al. Prediction of risk of distant recurrence using the 21-gene recurrence score in node-negative and node-positive postmenopausal patients with breast cancer treated with anastrozole or tamoxifen: a TransATAC study. J Clin Oncol. 2010; 28(11):1829-1834.
5. Habel LA, Shak S, Jacobs MK, et al. A population-based study of tumor gene expression and risk of breast cancer death among lymph node-negative patients. Breast Cancer Res. 2006; 8(3): R25.
6. Toi M, Iwata H, Yamanaka T, et al. Clinical significance of the 21-gene signature (Oncotype DX) in hormone receptor-positive early stage primary breast cancer in the Japanese population. Cancer. 2010; 116(13):3112-8.
7. Goldstein L, Gray R, et al. Prognostic utility of the 21-gene assay in hormone receptor–positive operable breast cancer compared with classical clinicopathologic features. J Clin Oncol. 2008; 26(25):4063-71.

 

Next: Clinical Utility and Health Economic Studies