Technology Evaluation
Center (TEC)

Gene Expression Profiling in Women with Lymph-Node-Positive Breast Cancer to Select Adjuvant Chemotherapy Treatment

Executive Summary

Background

For women with early stage breast cancer, adjuvant chemotherapy provides a significant proportional benefit, i.e., benefit that is the same regardless of prognosis, and regardless of hormonal treatment for hormone-receptor-positive tumors. However, a large proportion of these women would be disease free at 10 years without systemic therapy or with hormonal treatment alone; only a small proportion actually derive benefit from chemotherapy. Current risk classifiers do not accurately identify those early stage patients who are at low risk of recurrence; as a result, more patients are treated with chemotherapy than can benefit. Better predictors of baseline risk could help women who prefer to avoid the toxicity of chemotherapy, if assured that their risk is low, make better treatment decisions in consultation with their physicians.

Gene expression tests for improving the evaluation of recurrence risk in early stage breast cancer have been developed, evaluated, and made commercially available in the U.S. For example, Genomic Health, Inc. (Redwood City, CA) developed a 21-gene expression profile known as Oncotype DX™. Two major validation studies in women with early stage, lymph node-negative disease indicated that the summary test result (recurrence score or RS) provides significant recurrence risk information in addition to conventional criteria for individual patient risk classification. Additional evidence in a study of community-based patients with negative lymph nodes indicates that the 21-gene expression profile results are significantly associated with breast cancer death, and that RS high-risk patients clearly benefit from chemotherapy, whereas benefits for other RS categories are not statistically significant. More recently, Genomic Health has supported studies of the 21-gene expression profile in women with node-positive breast cancer.

Two other gene expression assays to predict distant recurrence after surgery are commercially available and have published validation data available. MammaPrint® (Agendia, Amsterdam, The Netherlands) evaluates the expression profile of 70 genes in fresh breast cancer tissue samples to assess a patients’ risk for distant metastasis (up to 10 years for patients younger than 61 years old, up to 5 years for patients 61 years or older). The test is cleared by the U.S. Food and Drug Administration (FDA) for patients with Stage I or Stage II lymph-node-negative breast cancer, with tumor size 5.0 cm or smaller and is indicated for use only as a prognostic marker along with other clinicopathological factors. The THEROS Breast Cancer IndexSM (bioTheranostics, San Diego, CA) is a simultaneous assessment of THEROS H/ISM (HOXB13:IL17BR, formerly Aviara H/ISM) and THEROS MGISM (Molecular Grade Index, formerly Aviara MGISM). According to the bioTheranostics website, the THEROS Breast Cancer Index “improves risk stratification in patients with estrogen receptor (ER)-positive, lymph node-negative breast cancer.”

Objective

This Assessment examines whether, compared to conventional risk assessment tools, the use of gene expression profiling improves outcomes when used to decide whether risk of recurrence is low enough to forego adjuvant chemotherapy for women with ER-positive, lymph-node-positive breast cancer. Because both MammaPrint and the THEROS Breast Cancer Index are intended for use in patients with lymph-node-negative breast cancer, only the 21-gene expression profile (Oncotype DX™) will be considered in this Assessment. (Note that an earlier TEC Assessment [2007; Vol. 23, No. 13] evaluated a similar question for all 3 assays for women with early stage, lymph-node-negative disease.)

Search Strategy

Published validation studies of the 21-gene expression profile (Oncotype DX™) obtained from the search for the previous TEC Assessment were checked for inclusion of patients with lymph-nodepositive disease. Published articles and abstracts posted on the Genomics Health website were also searched for relevance for this Assessment. Publications by authors of abstracts on studies of the 21-gene expression profile for patients with lymph-node-positive breast cancer presented at recent conferences were search for using PubMed’s Single Citation Matcher.

Selection Criteria

Included studies were full-length journal publications reporting on the use of the 21-gene expression profile (Oncotype DX™) to predict breast cancer recurrence for the purpose of identifying women with lymph-node-positive breast cancer who were unlikely to benefit from postoperative adjuvant chemotherapy.

Main Results

The major publication that addresses the questions asked in this Assessment is a report of a prospective evaluation of samples from the Southwest Oncology Group (SWOG) completed trial 8814, INT-0100. The patients in SWOG trial 8814 were postmenopausal women with axillary node-positive adenocarcinoma of the breast that was estrogen receptor (ER) and/or progesterone receptor (PgR)-positive. The trial arms of interest consisted of one arm in which postsurgery patients were treated only with tamoxifen and a comparison arm in which patients were treated with chemotherapy (cyclophosphamide, doxorubicin, and 5-fluorouracil) followed by tamoxifen. The purpose of the prospective sample evaluation was to evaluate the effect of the 21-gene RS on disease-free survival (DFS; primary outcome) and overall survival (OS; secondary outcome) by treatment group, to determine whether it identified those who might not benefit from anthracycline-based chemotherapy, despite the high recurrence risks associated with lymph-node positivity.

RS was prognostic in patients treated only with tamoxifen: 10-year DFS and OS outcomes differed by RS risk category (p=0.017 and 0.003, respectively). The “TAM” column in the Table shows the 10-year DFS for the low-risk RS category is 60%, for intermediate risk is 49%, and for high risk is 43%. In a Cox regression analysis, the hazard ratio (HR) for continuous RS was constant over time only for the first 5 years; RS was a significant prognostic factor (p=0.0002). Results were similar for overall survival. Missing was an analysis to show improved prognostic reclassification after classification of recurrence risk by standard clinical classifiers in the tamoxifen-only group.

When the 2 treatment arms were compared within each RS risk category for DFS, only patients in the high RS category significantly benefited from the addition of CAF to tamoxifen; similar results were reported for OS. In a stratified analysis of hazard ratios for disease-free survival, only the high RS group showed a significant benefit from the addition of chemotherapy (p=0.033). However, the low patient numbers and disease recurrence events, particularly in the low and intermediate RS risk categories of each treatment arm, resulted in wide confidence intervals (see Table) and likely low power for detecting significant differences.

The effect of the RS on treatment was not constant over time; an analysis of the first 5 years of follow-up was significant at p=0.029, but not after 5 years. In women with 1 to 3 positive nodes, low-risk RS patients have average 5-year recurrence risks at or below 10%, regardless of treatment. For women with 4 or more positive nodes, the low-risk RS 5-year recurrence risk average is closer to 20–25%, also regardless of treatment.

The interaction of RS with DFS remained significant after adjusting for age, race, tumor size, progesterone receptor status, grade, p53, and HER2; however, the interaction became nonsignificant (p=0.15) after adjusting for Allred-scored ER level (by immunohistochemistry).

Additional evidence in the form of single-arm trials suggested that patients with lymph-node-positive breast cancer who are treated with chemotherapy are more likely to respond if their RS is high than if it is low, but control arms for comparison were lacking.

Table.  Comparison of treatment arms for 10-yr DFS by RS category

Discussion

The statistical analysis in this study indicates that the 21-gene expression profile (Oncotype DX™) RS was prognostic in patients treated only with tamoxifen; however, whether the RS improves risk classification compared to standard clinical and histopathologic characteristics is unknown. The study results also suggest that in women with lymph-node-positive, ER-positive breast cancer, the 21-gene expression profile (Oncotype DX™) may help identify those women who are unlikely to benefit from the addition of CAF chemotherapy to their treatment regimen, despite their higher risk of recurrence due to positive nodes. However, the event rate was low, especially in the low-risk RS group, resulting in broad confidence intervals that included the possibility of benefit from chemotherapy.

Due to the lack of clear and sufficient information, there is a need for a second, confirmatory study. Under an award from the National Cancer Institute, the Fred Hutchinson Cancer Research Center will conduct a nationwide Phase III clinical trial to determine the predictive ability of the 21-gene expression profile to identify which patients with lymph-node-positive breast cancer will benefit from chemotherapy.

Based on the available evidence, the Blue Cross and Blue Shield Association Medical Advisory Panel made the following judgments about whether the 21-gene expression profile (Oncotype DX™) for selecting adjuvant chemotherapy in patients with lymph-node-positive breast cancer meets the Blue Cross and Blue Shield Association Technology Evaluation Center (TEC) criteria.

1. The technology must have final approval from the appropriate governmental regulatory bodies.

The 21-gene expression profile is available from only one laboratory and is not cleared by the U.S. Food and Drug Administration (FDA). Clinical laboratories may develop and validate tests inhouse (laboratory-developed tests or LDTs; previously called “home-brew”) and market them as a laboratory service; LDTs must meet the general regulatory standards of the Clinical Laboratory Improvement Act (CLIA). Laboratories offering LDTs must be licensed by CLIA for high-complexity testing. While the FDA has technical authority to regulate LDTs, to date there has been no active oversight with the exception of a more recently described category, “in vitro diagnostic multivariate index assay” (IVDMIA) devices. As of this writing, however, the Oncotype DX™ 21-gene expression profile has not been publicly classified by the FDA as an IVDMIA.

2. The scientific evidence must permit conclusions concerning the effect of the technology on health outcomes.

The available evidence does not permit conclusions regarding the use of the 21-gene expression profile for selecting adjuvant chemotherapy in patients with lymph-node-positive breast cancer. The evidence for not selecting chemotherapy for women with low RS values is based on low event rates and wide confidence intervals that include the possibility of benefit from chemotherapy. The results have not been confirmed in a second study.

3. The technology must improve the net health outcome; and

4. The technology must be as beneficial as any established alternatives.

The available evidence does not permit conclusions regarding the effect on health outcomes of using the 21-gene expression profile to select adjuvant chemotherapy in patients with lymphnode- positive breast cancer. Because the data allow for a possible benefit of chemotherapy in patients with low RS results, it is unknown if health outcomes would be improved, the same, or worse, if chemotherapy was withheld in these women.

5. The improvement must be attainable outside the investigational settings.

It has not yet been demonstrated whether use of the 21-gene expression profile for selecting adjuvant chemotherapy in patients with lymph-node-positive breast cancer improves health outcomes in the investigational setting. Therefore, it cannot be demonstrated whether improvement is attainable outside the investigational settings.

Based on the above, use of the 21-gene expression profile (Oncotype DX™) for selecting adjuvant chemotherapy in patients with lymph-node-positive breast cancer does not meet the TEC criteria.  

Full Study

25_01

Full studies are in PDF format. You will need Adobe Acrobat Reader to view all studies.
Download Adobe Acrobat Reader here.

adjuvant; Adjuvant! Online; AI; Amsterdam; analyte specific reagents; analyte-specific; analytic validity; aromatase inhibitor; ASCO; ASR; assay kit; breast cancer, Breast Cancer Gene Expression Ratio; estrogen receptor; cancer; cDNA; chemohormonal therapy; chemotherapy; clinical utility; clinical validity; complimentary; gene expression panel; gene signature; genetic test; genetic testing; Genomic Health; genomics; guideline; guidelines; home brew; homebrew; in vitro diagnostic multivariate index assay; IVDMIA; microarray; microarray; MINDACT; morbidity; mortality; multigene; multi-gene; National Comprehensive Cancer Network; National Surgical Adjuvant Breast and Bowel Project; NCCN; negative; neoadjuvant; node negative; node-negative; node positive; node-positive; NSABP; oncology; Oncotype DX; polychemotherapy; positive; prediction; prognosis; recurrence; recurrence score; RS; St. Gallen; SWOG: tamoxifen; TransATAC; TRANS-BIG; tumor grade; tumor makers; tumor size; tumor; validation; women???s health;