Medical Policy

 

Subject: Gene Expression Profiling for Managing Breast Cancer Treatment
Document #: GENE.00011 Publish Date:    12/27/2017
Status: Revised Last Review Date:    11/02/2017

Description/Scope

 

This document addresses the use of genetic profiling of breast tumors to predict breast cancer recurrence and response to therapy.

 

Position Statement

Medically Necessary:

Gene expression profiling with the Oncotype DX® Breast Recurrence Score, EndoPredict®, Prosigna Breast Cancer Prognostic Gene Signature Assay, or the Breast Cancer IndexSM as a technique for managing the treatment of breast cancer is considered medically necessary when all of the following criteria are met:

  1. Individual has had surgery and full pathological evaluation of the specimen has been completed; and
  2. Histology is ductal, lobular, mixed, or metaplastic; and
  3. Histology is not tubular or colloid (also referred to as mucinous); and
  4. Estrogen receptor positive (ER+), or progesterone receptor positive (PR+), or both; and
  5. HER2 (human epidermal growth factor receptor-2) receptor negative, based on any of the following:
    1. In situ hybridization (ISH) Testing:
      1. Single-probe average HER2 copy number less than 4.0 signals/cell, or
      2. Dual-probe HER2/CEP17 (chromosome enumeration probe 17) ratio less than 2.0 with an average HER2 copy number less than 4.0 signals/cell; or
    2. Immunohistochemistry (IHC) Testing:
      1. Zero, or
      2. 1+; and
  6. pN0 (node negative) or pN1mi with axillary lymph node micrometastasis less than or equal to 2 mm; and
  7. Any of the following:
    1. Tumor size 0.6-1.0 cm moderate/poorly differentiated; or
    2. Tumor size 0.6-1.0 cm and well-differentiated with any of the following unfavorable features: angiolymphatic invasion, or high nuclear grade, or high histologic grade; or
    3. Tumor greater than 1.0 cm and less than or equal to 5.0 cm; and
  8. Chemotherapy is a therapeutic option being considered by the individual and their provider; and
  9. No other breast cancer gene expression profiling assay has been conducted for the same tumor (for example a metastatic focus) or from more than one site when the primary tumor is multifocal.

Use of gene expression profiling with EndoPredict, Prosigna Breast Cancer Prognostic Gene Signature Assay, or the Breast Cancer Index as a genetic index used to assist in decisions of extending adjuvant hormonal therapy beyond 5 years of treatment is considered medically necessary when all of the following criteria are met:

  1. When criteria A through G above have been met; and
  2. When the Oncotype DX® Breast Recurrence Score was the initial gene expression profiling test used, and
  3. The individual is a candidate for additional hormonal or chemotherapy.

Not Medically Necessary: 

Gene expression profiling with the Oncotype DX Breast Recurrence Score, EndoPredict, Prosigna Breast Cancer Prognostic Gene Signature Assay, or the Breast Cancer Index as a technique of managing the treatment of breast cancer is considered not medically necessary when the criteria above have not been met.  

Investigational and Not Medically Necessary:

Gene expression profiling as a technique of managing the treatment of ductal carcinoma in situ (DCIS) is considered investigational and not medically necessary under all circumstances.  

Repeat gene expression profiling with the Oncotype DX Breast Recurrence Score, EndoPredict, Prosigna Breast Cancer Prognostic Gene Signature Assay, or the Breast Cancer Index for the same tumor (for example a metastatic focus) or from more than one site when the primary tumor is multifocal is considered investigational and not medically necessary.

Gene expression profiling as a technique of managing the treatment of breast cancer is considered investigational and not medically necessary when a gene profiling test other than the Oncotype DX Breast Recurrence Score, EndoPredict, Prosigna Breast Cancer Prognostic Gene Signature Assay, or the Breast Cancer Index is being used, including but not limited to:

  1. BluePrint (also referred to as "80-gene profile")
  2. Breast Cancer Gene Expression Ratio (also known as Theros H/ISM)
  3. BreastNext
  4. BreastOncPX
  5. BreastPRS
  6. Insight® DX Breast Cancer Profile
  7. MammaPrint® (also referred to as the "Amsterdam signature" or "70-gene signature")
  8. Mammostrat
  9. MammaTyper®
  10. NexCourse® Breast IHC4
  11. NuvoSelect eRx 200-Gene Assay
  12. Oncotype DX DCIS
  13. PAM50 Breast Cancer Intrinsic Classifier
  14. SYMPHONY Genomic Breast Cancer Profile
  15. TargetPrint®
  16. TheraPrint
  17. The 41-gene signature assay
  18. The 76-gene "Rotterdam signature" assay
  19. THEROS Breast Cancer IndexSM 
Rationale

The selection of individuals with breast cancer who may be candidates for chemotherapy is a complex and inexact science at this time.  The current tools available for recurrence risk assessment are limited and do not allow for great accuracy in the selection of appropriate individuals who would and would not benefit from treatment with chemotherapeutic agents.  More precise identification of these individuals could improve health outcomes through more appropriate chemotherapy use, mitigation of unnecessary treatment, and decreased adverse chemotherapy-related events.

The use of gene expression profiling assays has been proposed as a tool for identification of chemotherapy-appropriate individuals.  There are currently a wide array of different gene expression profiling assays either available or in various stages of development.  These tests are intended for use in identifying those individuals at low risk of recurrence for whom adjuvant chemotherapy can be avoided.  Several assays have been developed and validated using a variety of different genes and biomarkers.  However, the combined use of such assays has not been studied and it is unclear if such combined use provides additional clinical utility.

Bartlett (2016) evaluated the agreement between several different breast cancer assays using the samples of 313 subjects with estrogen receptor positive breast cancer as part of a larger study comparing women who had undergone treatment with standard care or test-directed by Oncotype DX assay.  This study compared the results of Oncotype DX, Prosigna, MammaPrint, NexCourse Breast, and conventional IHC4 assays.  Additionally, subtype classification by Blueprint, MammaTyper, and Prosigna were compared.  The authors reported that the five tests used showed “only modest agreement when dichotomizing results between high vs low/intermediate-risk were compared.  Concordant classification as either low/intermediate-risk or high-risk was noted in 119 (39.4%) tumors, and 183 (60.6%) were assigned to different risk categories by different tests.  Agreement between four of five tests was seen in 94 subjects (31.1%).  The subtype tests (BluePrint, MammaTyper, and Prosigna) assigned 59.5% to 62.4% of tumors to luminal A subtype.  However, only 121 (40.1%) were classified as luminal A by all three tests and only 58 (19.2%) were uniformly assigned as nonluminal A.  Discordant subtyping was observed in 123 (40.7%) tumors.  This study highlights significant variations in tumor risk and tumor sub-type results.

Oncotype DX

Oncotype DX was developed using the candidate gene method in which a relatively small number of genes known to be involved in breast cancer progression were selected and by analyzing expression of these genes in tumor specimens, a 21-gene signature predicting recurrence was developed.  Other assays were developed by analyzing gene expression of tumor specimens on large scale microarrays with thousands of gene transcripts, followed by pattern or cluster analysis to identify a much smaller gene signature that correlated with disease recurrence.  Two assays which have been compared, a 70-gene panel (MammaPrint or “Amsterdam signature”) and a 76-gene panel (“Rotterdam signature”), overlap by only three genes, due partly to the use of different microarray platforms in developing the panels.  More recent studies indicate that the two panels share 21 biological pathways, if not the same genes.

All these panels were developed using banked specimens from clinical trials or cohorts for which long-term outcomes were already known.  This is an efficient method for defining and establishing the clinical validity of the gene expression signatures.  Clinical validity for this application is defined as evidence supporting the ability of the panel to accurately predict outcomes such as disease recurrence, disease-free survival (DFS), or overall survival (OS).  Evidence of clinical validity has been reported in full-length journal publications for three panels; Oncotype DX, MammaPrint, and the Breast Cancer Gene Expression Ratio. 

Several retrospective studies have been published that support the clinical utility of Oncotype DX in individuals with node-negative breast cancer.  Paik (2006) used available, banked specimens from the randomized tamoxifen-positive chemotherapy-treated arms of National Surgical Adjuvant Breast and Bowel Project (NSABP) trial B-20 compared to the tamoxifen-only arm (samples also used as the test set to develop the assay) and correlated gene expression signatures to chemotherapy benefit.  The 424 selected subjects with high recurrence score (RS≥31) had an absolute increase in distant recurrence-free survival (DRFS) at 10 years of 27.6 ± 8% (mean ± standard error [SE]; p=0.01) compared to the tamoxifen-only group.  Subjects with low RS (<18) had little benefit from chemotherapy (absolute DRFS increase at 10 years, -1.1 ± 2.2%).  Interaction between chemotherapy and RS was significant at p<0.05.  These results suggest that the Oncotype DX RS is closely associated with the magnitude of the benefit from chemotherapy.

A second study by Habel (2006) described a retrospective case control study of 4964 subjects with node-negative invasive breast cancer.  In this study, cases (n=220) were defined as subjects who died of breast cancer and controls (n=570) were described as individually matched breast cancer subjects who were alive at the time of the study.  The results of the study found that the RS from Oncotype DX assay was associated with the risk of breast cancer-related death in estrogen receptor (ER) positive, tamoxifen-treated subjects.  At 10 years, the risk for death in this population was found to be 2.8% in low RS individuals, 10.7% in intermediate RS individuals and 15.5% in high RS individuals.  Individuals who were ER positive but not treated with tamoxifen had a risk of death of 6.2% in low RS individuals, 17.8% in intermediate RS individuals and 19.9% in high RS individuals.

In 2015, Sparano and colleagues published the results of a prospective cohort study involving 10,253 women age 18 to 75 years old.  Inclusion criteria were hormone receptor positive, HER2 negative (HER2-), node-negative breast cancer with tumor size between 1.1 and 5.0 cm of any grade or 0.6 to 1.0 for intermediate to nuclear grade (or both).  Using the Oncotype DX on this population, 1626 women had an RS result of 0-10, indicating a very low risk of recurrence.  This low-risk cohort prospectively received hormone therapy without chemotherapy.  The authors reported that invasive disease-free 5-year survival was 93.8% (95% confidence interval [CI], 92.4-94.9), freedom from recurrence of breast cancer at a distant site at 5 years was 99.3% (95% CI, 98.7-99.6), freedom from recurrence at 5 years was 98.7% (95% CI, 97.9-99.2) and OS at 5 years was 98.0% (95% CI, 97.1-98.6).  Overall recurrence rates were not significantly different when stratified by histological grade or age at diagnosis.  The authors concluded that their findings supported the use of the Oncotype DX test to spare the use of chemotherapy in subjects who would otherwise receive it on the basis of clinicopathologic features.

Gluz (2016) reported the 3-year results from the PlanB trial, in which Oncotype DX was prospectively used to define a subset of subjects to receive endocrine-only therapy.  The study enrolled 2568 subjects with node-positive or high-risk node and HER2- early breast cancer following surgical intervention.  No subjects had distant metastases detected.   The objective of the study was to compare the use of independent prospective central pathology review and assessment of immunohistochemistry markers vs. use of the Oncotype DX RS and local pathology.  The central laboratory pathologists were blinded.  In the overall study population, 18% were classified as low RS, 60.4% were intermediate RS, and 21.6% were high RS.  Based on their low RS status, 348 subjects had chemotherapy omitted from their treatment regimen.  There were 135 events reported during the reported trial period, with 73 (55.3%) occurring in the HER2 positive (HER2+) population, including 54 distant recurrences, 11 secondary neoplasms and local relapses, and 8 deaths without relapse.  Three year DFS was significantly poorer in the high-risk RS group vs. both the low- and intermediate-risk groups (91.9% vs. 97.8% and 97.4%, respectively, p<0.001).  Nodal status, central and local grade, the Ki-67 protein encoded by the MKI67 gene, estrogen receptor, progesterone receptor, tumor size, and RS were univariate prognostic factors for DFS; only nodal status, both central and local grade, and RS were independent multivariate factors.

Petkov and colleagues (2016) published the results of a large study involving 45,287 subjects with hormone receptor positive non-metastatic primary invasive breast cancer with available RS identified in the National Cancer Institute’s Surveillance Epidemiology and End Results (SEER) database.  Node negative status was reported for 40,134 subjects, 4691 had micrometastases or up to three positive nodes, and 642 had four or more positive nodes.  The Oncotype DX test was completed for 99.5% of all subjects with fewer than four positive nodes.  In subjects with 4-9 positive nodes, only 1% of subjects had Oncotype DX results.  The mean follow-up for the group with node-negative disease was 39 months vs. 30 months for those with node-positive disease.  A total of 38,568 subjects were included in the primary analysis of subjects with hormone receptor positive, HER2-, node-negative, non-metastatic disease who had RS results.  In this group, 21,023 subjects (55%) had RS results < 18, 14,494 subjects (38%) had RS results 18-30, and 3051 subjects (8%) had RS results ≥ 31.  The authors reported that breast cancer specific mortality (BCSM) was significantly associated with RS (p<0.001), with unadjusted 5-year estimates of 0.4%, 1.4%, and 4.4%, respectively.  A total of 4691 subjects with positive lymph nodes had RS results; 57% had RS < 18, 36% had RS results 18-30, and 7% had RS results ≥ 31.  Among this population, the 5-year BCSM was significantly different for the three RS groups (1.0%, 2.3%, and 14.3%, respectively; p<0.001).  The analysis was reported to have identified RS group as significantly prognostic (p<0.001) for 5-year BCSM, regardless of node status.  Five-year BCSM was 1.3% or lower for subjects with RS results < 18, regardless of nodal status and age group.  Low BCSM was also observed for subjects > 70 years of age with RS results 18-30.  However, for subjects with RS results ≥ 31, 5-year BCSM was substantially higher.  Subjects with node-positive disease with RS ≥ 31 had 5-year BCSM greater than 9.5%, regardless of age.  Higher RS was also associated with higher 5-year BCSM regardless of node status, race, or socioeconomic status.  Multiple linear regression modeling, controlling for a variety of factors, demonstrated that the 18-30 and ≥ 31 RS groups remain at significantly higher risk of BCSM (p<0.001).  RS results continued to be prognostic regardless of chemotherapy use (p=0.03).  These results support the conclusions that use of RS is a reliable prognostic tool for predicting 5-year BCSM.

The prognostic use of RS in genomically low-risk individuals who were clinically high-risk was reported by Nitz (2017).  The WSG-PlanB study involved 2642 subjects with unilateral primary invasive HER2-, node-positive or negative breast cancer randomized to receive treatment with either chemotherapy (n=1970) or endocrine-only therapy (n=348).  Mean follow-up was 55 months.  Compliance with treatment recommendations was 95.2% for the node-negative subjects and 75.2% in the node-positive subjects.  In hormone receptor positive subjects, 5-year DFS was higher in the RS < 11 and RS 12-25 groups vs. RS > 25 (93.6%, 94.3%, and 84.2% respectively; p<0.001 for comparisons with RS > 25).  In subjects with RS < 11 and treated with endocrine therapy alone, 5-year DFS was 94.2%, with no significant difference between node status groups.  In subjects with RS > 25, 5-year DFS was 61.7%.  The 5-year OS was 99% in endocrine-treated subjects with RS < 11 vs. 97% in those with RS 12-14, and 93% for those with RS > 25 (p<0.001).  The authors concluded that, “The excellent five-year outcomes in clinically high-risk, genomically low-risk (RS ≤ 11) pN0-1 patients without adjuvant chemotherapy support using RS with standardized pathology for treatment decisions in HR+ HER2 negative EBC.”

Stemmer (2017a and 2017b) published two studies reporting the outcomes of subjects, both node-negative and node-positive, who underwent RS-guided treatment for ER+ HER2- breast cancer.  The first report (2017a) involved 1801 ER+, HER2-, node-negative subjects with a mean follow-up of 6.2 years.  The authors reported that 48.9% of subjects had RS < 18, 40.7% had RS 18-30, and 10.4% had RS ≥ 31.  The lower RS groups has a higher proportion of grade 1 tumors and a lower proportion of grade 3 tumors compared to the higher RS groups.  Subjects judged to be at very low risk based on clinicopathologic characteristics were reported in all RS groups.  Adjuvant chemotherapy use was consistent with RS results with 1.4%, 23.7%, and 87.2% of subjects receiving chemotherapy in the < 18, 18-30 and the ≥ 31 groups, respectively.  The 5-year distant recurrence rates were significantly different between groups, with 0.8%, 3.0% and 8.6% reported, respectively.  In the RS ≥ 18 group, risk of distant recurrence increased with increasing tumor size.  The risk of 5-year breast cancer-related death was also significantly associated with RS group, with the risk reported to be 0.0%, 0.9% and 6.2%, respectively.  Multivariable regression modeling showed a significant association with RS and distant recurrence. 

The second Stemmer report (2017b) involved 709 subjects with ER+, HER2- subjects with 1-3 positive nodes with a mean follow-up of 5.9 years.  RS distribution was reported to be as follows: 53.4% of subjects had RS < 18; 36.4% had RS 18-30; and 10.2% had RS ≥ 31.  Adjuvant chemotherapy use was consistent with RS results with 7.1%, 39.5%, and 86.1% of subjects receiving chemotherapy in the < 18, 18-30 and the ≥ 31 groups, respectively.  The 5-year distant recurrence rates were significantly different between groups, with 3.2%, 6.3% and 16.9% reported, respectively.  Stratified by nodal status, 5-year distant recurrence rates were 1.2% for subjects with N1mi, 4.4 for those with 1 positive node, and 5.4% for 2-3 positive nodes.  The risk of 5-year breast cancer-related death was also significantly associated with RS group, with the risk reported to be 0.5%, 3.4% and 5.7%, respectively (p<0.001 between groups).  In the endocrine therapy only subjects (90.2% of subjects with RS < 18 and 59.3% with RS 18-30), the 5-year distant recurrence rates were 2.7% and 9.9%, respectively (p<0.001), and corresponding BCSM was 0.6% and 5.1% (p=0.002).  The 5-year distant recurrence rate in subjects with RS ≤ 31 was stratified by adjuvant chemotherapy use.  In subjects with RS < 18, the recurrence rate was 7.7% for subjects who received chemotherapy (n=27) vs. 2.9% in those who did not (n=352; p=0.245).  In subjects whose RS was 18-30, the recurrence rate in the chemotherapy-treated subjects (n=102) was significantly lower than untreated subjects patients (n=156; 1.0% vs. 9.7%, p=0.019).  The authors surmised that this difference was due to the subgroup of RS 26-30 and not from the 18-25 subgroup (p=0.017 and 0.058, respectively).  As with the study in node-negative subjects, a multivariable regression modeling showed a significant association with RS and distant recurrence.

The use of the Oncotype DX test for individuals with node-positive breast cancer has been described in a study by Eiermann and colleagues (2012).  This study involved 379 subjects, 244 who were node-negative (N0) and 122 who were node-positive (N+), and evaluated how the Oncotype DX test influenced treatment decisions.  The results showed that treatment recommendations changed in 33% of all subjects (N0 30%; N+ 39%).  In 39% of N0 and 37% of N+ subjects with an initial recommendation for chemoendocrine therapy, the post-RS recommendation changed to endocrine therapy.  In 22% of N0 and 39% of N+ subjects with an initial recommendation for endocrine therapy only, treatment recommendations changed to combined chemoendocrine therapy.  Overall, 33% (N0 29%; N+ 38%) fewer subjects actually received chemotherapy as compared with subjects recommended for chemotherapy pre-test. This study indicates that there was a positive result from the use of the Oncotype DX test in impacting treatment recommendations in individuals with node-positive breast cancer. However, no data has yet been presented regarding health outcomes of Oncotype DX-directed treatment decisions for node-positive individuals.  Prognosis in node-positive individuals who are classified as low-risk remains poor with 10-year DFS of 60% and OS of 77%, respectively.  Therefore, such individuals should still be considered candidates for chemotherapy to decrease their risk of recurrence (Albain, 2010).  Further evaluation in this population is needed before treatment decisions in node-positive individuals should be informed by Oncotype DX.

The National Comprehensive Cancer Network (NCCN) Practice Guidelines in Oncology: Breast Cancer (V.2.2017) provides a framework in which this test may be used.  The recommendations made in the NCCN Practice Guideline regarding the 21-gene RT-PCR assay (Oncotype DX test) are based upon a Category of Evidence and Consensus 2A, which is defined as follows: “Based upon lower-level evidence, there is uniform NCCN consensus that the intervention is appropriate.”

Review of the literature indicates that data addressing the use of the Oncotype DX test for tumors greater than 4.0 cm is sparse.  In the major trials available to date, including the NSABP B-14 and B-20 studies used in development of the Oncotype DX algorithm as reported, more than 95% of all participants had tumors less than or equal to 4.0 cm (Fisher, 1994; Fisher, 1997; Habel, 2006; Paik 2004; Paik 2006).

Repeat testing of either a single sample or of a new sample from the same individual using the Oncotype DX test has not been evaluated in the scientific literature, and there are no peer-reviewed published studies available in the National Library of Medicine’s PubMed database that address this issue.  Additionally, Oncotype DX testing of multiple tumor sites in a single individual has not been addressed in the scientific literature.  There are no studies available in PubMed that refer to this issue.  Further evidence is needed to fully address the impact of repeat testing or the testing of multiple sites.  Unfortunately, it is not unknown for an individual to experience two separate and unrelated cases of breast cancer in their lifetime.  Under such circumstances, use of the Oncotype DX test to evaluate each separate case would not be deemed a repeat test, but separate tests.

Oncotype DX DCIS

The use of a modified version of the Oncotype DX test has been proposed for guiding treatment decisions in individuals with ductal carcinoma in situ (DCIS).  The DCIS variation of the Oncotype DX test, named the Oncotype DX DCIS test, relies on the expression of fewer genes than the original Oncotype DX test (12 vs 21), and the modifications are based on proprietary algorithms.  This test is scaled as a continuous variable from 1 to 100, with low risk defined as Oncotype DX DCIS Score (DS) less than 39, intermediate risk defined as DS from 39-54, and high risk defined as greater than or equal to 55.  The prospective use of this test has been addressed in a published article by Solin (2013).  This study involved 327 tissue samples from subjects who participated in the ECOG E5194 study, which was a nonrandomized prospective trial designed to evaluate the use of surgical excision without radiation for women with DCIS.  The authors reported that the DCIS Score was significantly associated with the development of an ipsilateral breast event (IBE), defined as local recurrence of DCIS or invasive carcinoma in the ipsilateral breast, when adjusted for tamoxifen use (hazard ratio [HR]=2.31; p=0.02).  This risk remained almost unchanged when the adjustment for tamoxifen use was removed (HR=2.38; p=0.01).  For invasive carcinoma alone, the HR was 3.68 (p=0.01).  In a multivariate analysis, DCIS Score, tumor size, and menopausal status were all independently and significantly associated with developing an IBE (p≤0.02 for all).  Using the DCIS Score to stratify risk into three pre-specified risk groups (low, intermediate, and high), the 10-year risk of an IBE was 10.6%, 26.7%, and 25.9%, respectively (log rank p=0.006).  The corresponding 10-year rates for developing invasive cancer alone were 3.7%, 12.3%, and 19.2%, respectively (log rank p=0.003).  No significant association was found between the DCIS Score and the risk of developing a DCIS or invasive cancer in the contralateral breast.  No association was reported for the standard Oncotype DX Recurrence Score and the risk of developing a DCIS or invasive cancer in either breast.  The results of this study demonstrate that while the use of the DCIS Score may help predict the risk of ipsilateral IBE in women with DCIS, this study was small and retrospective, and did not demonstrate any change in clinical outcomes as a result of care guided by the Oncotype DX DCIS test. 

A prospective case series study (Alvarado, 2015) involving 115 subjects with DCIS who were eligible for breast-conserving therapy was conducted to study the impact of Oncotype DX DCIS scoring on treatment recommendations for radiation therapy following breast conserving surgery (BCS).  Subjects with invasive carcinoma and planned mastectomy were excluded.  The median size of DCIS was 8 mm (range 1-115 mm).  Pre-assay median physician estimate of 10-year local recurrence risk was 20% (range: 6-60%) for any DCIS or invasive cancer and 10% for invasive cancer only (range: 3-25%).  Post-assay estimates were 16% (range: 5-25%) for any DCIS or invasive cancer and 7% (range: 2-25%) for invasive cancer only.  Pre-assay, 73% of subjects had radiotherapy (RT) recommendations vs. 59.1% post-assay (p=0.008).  This study demonstrated that the DCIS Score led to statistically significant changes in physician treatment choice.  This study was not an assessment of whether the decision was made appropriately, as no data was presented relating to the impact on health outcomes as a result of treatment guided by the results of the Oncotype DX DCIS test.

Rakovitch and others (2015) published the results of a retrospective study involving tumor block samples from 718 subjects with DCIS treated with BCS, 571 (79.5%) of whom had negative margins.  Median follow-up was 9.6 years.  In the primary pre-specified analysis, the DCIS Score was associated with any local recurrence (DCIS or invasive cancer) in ER+ individuals (HR=2.26; p<0.001) and in all subjects regardless of ER status (HR=2.15; p<0.001).  The DCIS Score provided independent information on local recurrence risk beyond clinical and pathologic variables including size, age, grade, necrosis, multifocality, and subtype (adjusted HR=1.68; p=0.02).  The DCIS score was associated with invasive local recurrence (HR=1.78; p=0.04) and DCIS local recurrence (HR=2.43; p=0.005).  The authors concluded that the DCIS Score independently predicts and quantifies individualized recurrence risk in a population of individuals with pure DCIS treated by BCS alone.  This report does not provide any data related to the impact on health outcomes as a result of treatment guided by the results of the Oncotype DX DCIS test.

This same group reported another study involving tumor blocks from 1260 subjects with DCIS treated with BCS; 571 treated with BCS alone and 689 treated with BCS plus RT (Rakovitch, 2017).  All samples came from tumor samples greater than 2 mm.  Median follow-up for subjects from whom the samples were available was 9.4 years.  At baseline the BCS+RT group had significantly more adverse features vs. the BCS alone group including age less than 50, high nuclear grade and presence of comedo-necrosis (p<0.001 for all).  The 10-year cumulative risk of local recurrence in the BCS alone group was 19.2% vs. 12.7% in the BCS+RT group.  In the BCS alone group, the factors associated with local recurrence included DS, multifocality, age at diagnosis and subtype.  In the BCS+RT group subjects with clear margins, the factors associated with local recurrence included multifocality and tumor size greater than 1 cm.  DS was associated with local recurrence in this group when adjusting for these factors (p<0.001).  The HR for subjects with high-risk DS was 2.53 when compared to subjects with low-risk DS (p<0.001).  For subjects with intermediate-risk DS, the HR was 1.62 vs. the low-risk DS group (p=0.11).  The 10-year risk of local recurrence was 20.5% in the high-risk group, 13.6% in the intermediate group and 7.5% in the low-risk group (p<0.001).  When the BCS alone and BCS+RT groups were pooled for analysis, the authors reported that age at diagnosis less than 50, tumor size greater than 1 cm, multifocality, and the administration of RT were associated with local recurrence.  When adjusting for these factors, DS was still a significant predictor of local recurrence (HR=1.97, p<0.001).  Furthermore, the HR for the high-risk and intermediate groups was significantly higher than the low-risk group (1.77 and 1.73, respectively).  No interactions were found between DS and RT (p=0.44).  When adjusted for propensity score, subjects with low-risk DS had a 10-year risk of local recurrence of 16% after BCS alone and 9.5% after BCS+RT.  The 10-year risk of invasive local recurrence was 9.7% and 6.8%, respectively.  For the high-risk DS group, when adjusted for propensity score, the 10-year risk of local recurrence was 32.7% in the BCS alone group and 20.0% in the BCS+RT group.  The risk of invasive local recurrence was 17% and 11/9%, respectively.

MammaPrint    

The translational research network of the Breast International Group (TRANS-BIG) published validation studies for the MammaPrint 70-gene breast cancer recurrence assay (Buyse, 2006; Desmet, 2007; Rutgers, 2011).  In August 2016, this group published the results of the Microarray for Node Negative Disease may Avoid Chemotherapy Trial (MINDACT) (Cardoso, 2016).  This study was designed to examine the additional clinical utility of the 70-gene signature test (MammaPrint) to standard clinical–pathological criteria in selecting subjects for adjuvant chemotherapy.  In a prospective, randomized, phase III study, 6693 women with early-stage breast cancer had their genomic risk and clinical risk determined using the MammaPrint and a modified version of Adjuvant! Online, respectively.  Women at low clinical and genomic risk did not receive chemotherapy (2745 subjects, 41.0%), whereas those at high clinical and high genomic risk did receive chemotherapy (1806 subjects, 27.0%).  Subjects with discordant risk results were randomized to determine the use of chemotherapy using either the clinical or genomic risk.  The primary goal was to assess whether the lower boundary of the 95% confidence interval for the rate of 5-year survival without distant metastasis would be 92% (that is, the non-inferiority boundary) or higher in subjects with high clinical risk and a low-risk gene expression profile who did not receive chemotherapy.  A total of 1550 subjects (23.2%) were determined to be at high clinical risk and low genomic risk.  At 5 years, the rate of survival without distant metastasis in this group was 94.7% (95% CI, 92.5 to 96.2) among those not receiving chemotherapy, higher than the non-inferiority boundary and therefore achieving the primary goal.  The absolute difference in survival rate between subjects with high clinical risk and low genomic risk who received chemotherapy vs. those who did not receive chemotherapy was 1.5 %, with the rate being lower without chemotherapy.  Similar rates of survival without distant metastasis were reported in the subgroup of subjects who had ER positive, HER2-, and either node-negative or node-positive disease.  In a secondary analysis, there was no advantage in directing therapy on the basis of genomic risk among subjects at low clinical risk and high genomic risk, since these subjects had no benefit from the use of adjuvant chemotherapy [those receiving chemotherapy, 5-year rate of survival without distant metastasis of 95.8% (95% CI, 92.9 to 97.6), compared to those without chemotherapy with a rate of 95.0% (95% CI, 91.8 to 97.0%), adjusted HR of 1.17 (95% CI, 0.59 to 2.28; p=0.66)], and there was no significant difference between the chemotherapy group and the no-chemotherapy group with respect to DFS and OS.  The authors concluded that for women with early-stage breast cancer who were at high clinical risk and low genomic risk for recurrence, avoiding chemotherapy on the basis of MammaPrint testing led to a 5-year rate of survival without distant metastasis that was 1.5 percentage points lower than the rate with chemotherapy and that given these findings, when using MammaPrint results, approximately 46% of women might not require chemotherapy.

Van de Vijver (2002) published one of the first studies describing the use of the MammaPrint test using a consecutive series of frozen tumor samples from 295 subjects with primary stage I or II breast carcinoma.  All subjects were younger than 53 years of age; 151 had positive lymph nodes, and 144 had negative lymph nodes.  The 70-gene assay was used to predict poor or good prognosis.  The authors reported 180 subjects had a poor prognostic signature and 115 had a good prognostic signature.  The mean overall 10 year survival rate was 54.6 ± 4.4% and 94.5 ± 2.6%, respectively.  At 10 years, the probability of remaining free of distant metastases was 50.6 ± 4.5% in the group with a poor prognosis signature and 85.2 ± 4.3% in the group with a good prognosis signature.  The estimated HR for distant metastases in the group with a poor prognosis signature, vs. the good prognosis signature group was 5.1 (p<0.001).  This ratio remained significant when the groups were analyzed according to lymph node status.  Multivariable Cox regression analysis showed that the prognosis profile was a strong independent factor in predicting disease outcome.

In 2004, Drukker and colleagues did a follow-up study using the population originally described by Van de Vijver (2002).  The median follow-up for this series was extended to 18.5 years.  A significant difference was reported in long-term distant metastasis-free survival (DMFS) for the subjects with a low, and a high-risk 70-gene signature (DMFS p<0.0001), as well as separately for node-negative (DMFS p<0.0001) and node-positive subjects (DMFS p=0.0004).  The 25-year HRs for all subjects for DMFS and OS were 3.1 (95% CI, 2.02-4.86) and 2.9 (95% CI, 1.90-4.28), respectively.  The HRs for DMFS and OS were largest in the first 5 years after diagnosis: 9.6 (95% CI, 4.2-22.1) and 11.3 (95% CI, 3.5-36.4), respectively.  The 25-year HRs in the subgroup of node-negative subjects for DMFS and OS were 4.57 (95% CI, 2.31-9.04) and 4.73 (95% CI, 2.46-9.07), respectively, and for node-positive subjects for DMFS and OS were 2.24 (95% CI, 1.25-4.00) and 1.83 (95% CI, 1.07-3.11), respectively.  The authors concluded that the 70-gene signature remains prognostic at longer follow-up in subjects < 53 years of age with stage I and II breast cancer and that the 70-gene signature’s strongest prognostic power is seen in the first 5 years after diagnosis.

Mook and colleagues describe a pair of trials evaluating the prognostic ability of the MammaPrint.  The first study involved retrospective data and tumor samples from 241 subjects with node-positive breast cancer (Mook, 2009).  The 10-year DMFS and breast cancer specific survival (BCSS) probabilities were reported to be 91% and 96% respectively for the good prognosis group and 76% for both DMFS and BCSS for the poor prognosis group.  These results were compared to those obtained using Adjuvant! Online (AOL) and judged to be superior in predictive value.  The second study involved tumor samples and retrospective clinical data from 184 subjects between the ages of 55 and 70 previously treated for breast cancer (Mook, 2010).  The authors report that the BCSS at 5 years was 99% for the good prognosis signature versus 80% for the poor prognosis signature group (p=0.036).  They concluded that MammaPrint results were a significant and independent predictor of BCSS during the first 5 years of follow-up.  The other two studies evaluated the use of the MammaPrint test in subjects with node-negative disease (Bueno-de-Mesquita, 2007, 2009).  In both of these studies, MammaPrint prognostic index results were compared to other commonly used clinicopathologic risk indexes, and both reported favorable results.  None of these studies evaluated the impact of the MammaPrint test on overall survival, nor on the avoidance of unnecessary treatment with chemotherapy.  Finally, Knauer and others (2010) reported the results of a meta-analysis study involving the data from previous studies.  The dataset included data from 541 subjects who had received BCS or mastectomy with sentinel node biopsy or axillary lymph node dissection followed by RT.  For this study, data for all subjects who received endocrine therapy (ET) alone or ET plus adjuvant chemotherapy (ET+CT) was reanalyzed for benefit of adjuvant CT using groups classified as good or poor prognosis based upon MammaPrint testing as a framework.  Additionally, time-to-event data was gathered and analyzed from the pooled data.  The authors reported that BCSS and distant DFS were significantly better in subjects classified as having a good prognosis compared to those classified as poor prognosis (p<0.01).  Furthermore, the data showed that subjects in the poor prognosis group who had received ET+CT had a significantly longer BCSS and distant DFS when compared to those in the good prognosis group (BCSS 94% vs. 81%, p<0.01; distant DFS 88% vs. 76%, p<0.01).  No differences between the ET and EC+CT groups were found in relation to the benefit of CT in the good prognosis group.

Saghatchian (2012) described a study in which samples from 173 subjects with breast cancer and 4-9 positive nodes were evaluated with the MammaPrint test.  The study was proposed to evaluate the prognostic values of this test in this high-risk population.  Seventy (40%) of the samples were classified as low-risk and 103 (60%) as high-risk.  In the low-risk category, the 5-year OS was 97% vs. 76% for the high-risk group (p<0.01).  Distant metastasis-free survival at 5 years was reported to be 87% for low-risk subjects and 63% for high-risk subjects (p<0.01).  The authors note that in the Luminal A subgroup, the genomic profile was the only independent risk factor for distant metastasis (DM) and breast cancer (BC) specific death.  

Drukker and others provided data from a prospective evaluation of the MammaPrint test in the microarRAy-prognoSTics-in-breast-cancER (RASTER) study (2013).  This study involved 427 subjects with unilateral, unifocal, primary operable invasive adenocarcinoma of the breast with no positive nodes.  ER status was positive in 342 (80%) of subjects and negative in 85 (20%).  PR status was positive in 293 (69%), negative in 133 (31%), and unknown in 1 subject.  HER2 status was positive in 48 (11%) subjects, negative in 358 (84%) subjects and unknown in 21 (4%) subjects.  When comparing the results of the MammaPrint test to AOL risk prediction, there was 38% (161/427) discordance.  Most were seen in the signature-identified low-risk group and in the AOL high-risk identified group (124/427, 29%).  The rest were in the signature-identified high-risk group and in the AOL low-risk identified group (37/427, 9%).  Five-year distant recurrence-free survival was 98.4% in subjects with signature-identified low-risk / AOL high-risk (n=124), of which 76% (n=94) had not received adjuvant chemotherapy.  The group that had not received adjuvant chemotherapy had a 5-year distant recurrence-free survival of 98.9%.  The group that did not receive any systemic therapy (chemotherapy nor endocrine therapy) (n=70) had a 5-year distant recurrence-free survival of 100%.  No significant difference (p=0.29) was seen between systemically untreated subjects with a concordant low-risk assessment and subjects with a 70-gene signature low-risk result even with a high-risk assessment by AOL.  The authors conclude that 5-year distant recurrence-free survival probabilities demonstrated significant additional prognostic value of MammaPrint results to clinicopathological risk estimations such as AOL.  

In a follow-up to the RASTER study, Drukker (2014) evaluated the addition of MammaPrint data to several common risk estimation methods in optimizing outcome predictions and treatment decisions in individuals with early stage, node-negative breast cancer.  The risk estimation tools used in this study included Adjuvant! Online version 8.0 (AOL), Nottingham Prognostic Index (NPI), St. Gallen (2003), the Dutch National guidelines (CBO 2004 and NABON 2012), and PREDICT plus.  For the use of the AOL, high-risk was defined as a predicted 10-year survival of less than 90%.  On the NPI, moderate or high-risk was defined as a score greater than 3.4.  The authors reported that at a median follow-up time of 61.6 months, 24 distant recurrence-free interval (DRFI) events occurred, including the death of 11 subjects.  The 5-year DRFI probabilities using the MammaPrint tool for the 219 low-risk and 208 high-risk subjects were 97.0% and 91.7%, respectively (p=0.03).  Reporting concordance statistics in the form of a c-index measure where 0.5 equates no discrimination and 1.0 denotes perfect discrimination, the authors reported that the addition of the MammaPrint data to the risk estimation tools produced borderline improvement in most cases.  Following the addition of the MammaPrint data, the c-index rose to 0.638 for the NPI (p=0.05), to 0.639 for the 2004 Dutch national guidelines (p=0.04), to 0.662 for the PREDICT plus tool (p=0.27), and to 0.644 for the 2012 Dutch guidelines (p=0.05).  The lowest c-index measure was for the AOL, which went from 0.532 to 0.619 before and after the addition of the MammaPrint data (p=0.03).  Discordance between clinical risk estimations and the MammaPrint tool occurred 37% of the time for AOL, 27% for NPI, 39% for St. Gallen, 30% for the 2004 Dutch guidelines, 39% for the 2012 Dutch guidelines and 25% for the PREDICT plus tool.  The most discordant measures were in subjects classified as MammaPrint low-risk and clinically determined as high-risk.  When the MammaPrint data were factored in, fewer subjects were eligible to receive adjuvant chemotherapy with AOL (-20%), St. Gallen (-24%), 2004 Dutch guidelines (-6%), and 2012 Dutch guideline (-22%).  Conversely, with addition of the MammaPrint data more subjects were eligible for adjuvant chemotherapy with the NPI (7%) and PREDICT plus (2%) tools.  This study demonstrates that use of the MammaPrint tool may have impact when combined with some commonly used clinical risk assessment tools in individuals with early stage breast cancer.  However, the data indicates that the benefit of addition of the MammaPrint tool in conjunction with AOL is low. 

Kok and colleagues (2012) evaluated the use of the MammaPrint test in conjunction with ER and PR levels for the prediction of breast cancer outcomes.  This study evaluated three separate categories of subjects: (1) Those that received adjuvant tamoxifen (n=121, 81% node-positive); (2) Those that did not receive tamoxifen (n=151, 10% node-positive); and (3) Those that received tamoxifen for metastatic disease (n=92).  In subjects treated with adjuvant tamoxifen, both the MammaPrint as well as the endocrine response were associated with breast cancer specific survival.  In subjects treated with tamoxifen for metastatic disease, combined analysis of MammaPrint and ER/PR revealed additional value (p=0.013).  For those who did not receive tamoxifen, only MammaPrint was associated with outcome.  The authors concluded that in the series analyzed, MammaPrint was mainly a prognostic factor, while ER and PR levels were mainly associated with outcome after tamoxifen.

In 2017, Esserman and others reported on a study evaluating the “ultralow-risk threshold” of the MammaPrint test to determine 20-year risk of recurrence in 652 subjects enrolled in the STO-3 trial who have not has any breast cancer-related deaths within 15 years of initial treatment.  All subjects were postmenopausal with node-negative breast cancer and tumors ≤ 3 cm who had received either treatment with tamoxifen (n=313) or no endocrine therapy (n=339).  Another 15% (n=98) were classified as ultralow-risk.  Additionally, 538 were ER+, 369 were PR+, 8 were HER2+, and 178 has Ki67 ≥ 15%. Tumor grade 1 was reported in 19% of subjects, grade 2 in 58%, and grade 3 in 23%.  MammaPrint scored 42% (n=275) of subjects as high-risk and 58% (n=377) as low-risk.  A statistically significant difference was reported between groups (p<0.001), with low-risk subjects having > 95% 5-year breast cancer survival.  At 20 years, women with 70-gene high-risk and low-risk tumors, but not ultralow-risk tumors, had a significantly higher risk of disease-specific death compared with ultralow-risk subjects (HR=4.73 and 4.54, respectively).  No deaths were reported for the ultralow-risk subjects treated with tamoxifen at 15 years, and the 20-year disease-specific survival rate was 97%.  Untreated ultralow-risk subjects had a survival rate was 94%.  All ultralow risk tumors were HE+, HER2-, luminal subtype.  The authors concluded that, “The ultralow-risk threshold of the 70-gene MammaPrint assay can identify patients whose long-term systemic risk of death from breast cancer after surgery alone is exceedingly low.”

Further data is needed to demonstrate the potential benefits of MammaPrint testing on health outcomes in women with breast cancer.

BluePrint

The BluePrint test is intended as an adjunctive test to the MammaPrint test.  It is proposed to further stratify Luminal-type cancers and improve recurrence risk stratification.  Currently, the only peer-reviewed published evidence addressing the use of the BluePrint test is a retrospective study by Glück (2013).  The study involved 437 tissue samples from subjects from four different studies evaluating pathologic complete response (pCR) after neoadjuvant chemotherapy in individuals with early stage breast cancer using BluePrint and MammaPrint tests vs. clinical subtyping with immunohistochemistry/fluorescence in situ hybridization (IHC/FISH) for the determination of estrogen receptor, progesterone receptor, and human epidermal growth factor receptor-2 (HER2) status.  It was reported that pCR rate differed substantially among BluePrint molecular subgroups: 6% in Luminal A-type, 10% in Luminal B-type, 47% in HER2-type, and 37% in Basal-type samples.  The results indicated that for samples stratified as Luminal A-type samples by the BluePrint test, the pCR rate provided no prognostic information, suggesting these subjects may not benefit from chemotherapy.  This study is hampered by several methodological flaws, including differences in the definition of positive HER2 results, different treatment regimens and different testing methodologies between study sites.  Further data is needed from well conducted, prospective, large-scale randomized controlled trials (RCTs) to evaluate the clinical utility of the BluePrint test.

Breast Cancer Gene Expression Ratio

The current available data addressing the Breast Cancer Gene Expression Ratio test is limited to one retrospective study (Goetz; 2006).  In this study, tumor blocks from 206 women with breast cancer were examined for HOXB13/IL-17BR expression ratio (H:I ratio).  They found that in lymph node-positive subjects (n=86) the H:I ratio was not associated with relapse or survival.  However, in node-negative subjects (n=103) a high H:I ratio was associated with significantly worse disease-free survival, relapse-free survival, and overall survival.  Again, while the results of this study are promising, they are insufficient to allow generalized conclusions on how this test would perform in other populations.

Rotterdam 76-gene assay and 41-gene assay

Evidence addressing the use of the Rotterdam 76-gene assay, as well as the 41-gene assay, is limited by insufficient data.  Further data regarding these tests is required to properly assess their clinical utility.

Insight DX Breast Cancer Profile

A study published by Linke and colleagues in 2006 describes a retrospective case series study of 324 subjects with breast cancer.  This study used an early version of the Insight DX Breast Cancer Profile test.  The results of this test showed promise in being able to predict outcomes in tamoxifen-treated individuals with breast cancer.  However, this is the only peer-reviewed published study addressing this test, and thus there has been no independent confirmation that this test prospectively results in incrementally improved care and outcomes.

Mammostrat

One published study on the Mammostrat by Ring and others (2006) discusses a single validation study for this test, but no data has been made available regarding the impact of this test on clinical outcomes.  In 2012, Bartlett and others published the results of a study evaluating the efficacy of the Mammostrat test in the Tamoxifen versus Exemestane Adjuvant Multicenter (TEAM) trial.  The authors tested 4598 pathology blocks from TEAM participants, who were node-positive in 47% of subjects and in whom 36% were treated with adjuvant chemotherapy, and reported on 3837 that were successfully scored.  In the 1226 (31.9%) subjects that were both node-negative and did not receive chemotherapy, the Mammostrat test was a significant prognostic factor for distant relapse-free survival (p=0.004).  Subjects with moderate or high scores were reported to be 58% and 159% more likely to experience distant relapse that those with low Mammostrat scores.  Similarly, Mammostrat results were an independent factor in multivariate analysis for DFS in these populations (p=0.038).  In the sample of subjects treated without chemotherapy (n=2559), multivariate analysis found that Mammostrat score remained an independent predictor of distant relapse-free survival risk (p<0.001), with a 45% and 75% increase in recurrence risk for medium and high-risk scores, respectively, compared with subjects with low-risk scores.  However, for DFS, no significant benefit from Mammostrat was seen (p=0.085).  When a multivariate analysis was conducted in the total study population, analyses adjusted for conventional prognostic factors (i.e., nodal status, grade, size, age, treatment, HER2, and quantitative PR and ER), the Mammostrat score remained an independent predictor of distant relapse-free survival risk (P for trend <0.001) with a 50% and 91% increase in risk of recurrence for medium and high-risk scores, respectively compared with subjects with low-risk scores.  In a similar analysis for DFS, significant additional prognostic value of the Mammostrat score alongside conventional markers was found (P for trend <0.001).  The results from this trial are promising, but this is only an initial report of the use of the Mammostrat test.  Further studies seeking evidence addressing the clinical utility of this test are warranted.

PAM50 Breast Cancer Intrinsic Classifier

The PAM50 Breast Cancer Intrinsic Classifier involves the analysis of a set of 50 genes to classify individuals into four separate subtypes based on gene expression profiles: Luminal A, Luminal B, HER2-enriched, and Basal-type.  These subgroups, in addition to clinicopathologic data, have been proposed as a way to predict clinical outcomes in individuals with breast cancer.

A small comparative study by Kelly and others in 2012 described the agreement between the Oncotype DX test and the PAM50 Breast Cancer Intrinsic Classifier.  The PAM50 assay classifies four tumor types: Luminal A, Luminal B, HER2 enriched, and Basal-type.  The study utilized archived tumor block samples from 108 subjects who had previously received Oncotype DX testing and treatment.  Samples from the tumor blocks were evaluated with the PAM50 assay and compared with the Oncotype DX RS for each subject.  All subjects were ER+ and HER-; 96% had node-negative disease (n=102).  Most subjects (71%) did not receive chemotherapy but did receive adjuvant hormone therapy (94%).  The authors reported that 103 (95%) were classified as Luminal A (n=76) or Luminal B (n=27).  Ninety-two percent (n=98) had a low (n=59) or intermediate (n=39) RS.  Among Luminal A cancers, 70% had low RS (n=53) and the remainder (n=23) had an intermediate RS.  Among Luminal B cancers, 9 were high (33%) and 13 were intermediate (48%) by the RS.  Almost all cancers with a high RS were classified as Luminal B (90%, n=9).  One high RS cancer was identified as Basal-type and had low estrogen receptor/estrogen receptor gene 1 (ER/ESR1) and low human epidermal growth factor receptor (HER2) expression by quantitative polymerase chain reaction in both assays.  The majority of low RS cases were Luminal A (83%, n=53).  Importantly, half of the intermediate RS cancers were re-categorized as low-risk Luminal A subtype by PAM50.  The authors conclude that “There is good agreement between the two assays for high (i.e., Luminal B or RS > 31) and low (i.e., Luminal B or RS < 18) prognostic risk assignment but PAM50 assigns more individuals to the low risk category.”  Additional data are required to fully evaluate the clinical utility of the PAM50 assay.

Prosigna Breast Cancer Prognostic Gene Signature Assay

Another assay using the PAM50, 50 gene set is the Prosigna Breast Cancer Prognostic Gene Signature Assay.  This assay uses the NanoString nCounter® device to assess the PAM50 gene subtype data.  In addition to the PAM50 data, an additional algorithm using clinical information is used to derive an overall Risk of Recurrence (ROR) Score.  The ROR scores range from 0 to 100 and further stratify individuals into 1 of 3 risk categories, with scores of 0 to 40 assigned “low” risk, scores of 41 to 60 “intermediate” risk, and scores of 61 to 100 “high” risk. The available evidence regarding the use of the Prosigna test is limited to a few studies.

Dowsett and colleagues (2013) reported the results of a nonrandomized comparative study using banked samples from 1017 subjects involved in the ATAC (Anastrozole, Tamoxifen, Alone or in Combination) study.  The prognostic power of the Oncotype DX RS, ROR, or IHC4 (immunohistochemical prognostic model) was compared to that of a clinical treatment score derived from immunohistochemical assessment of ER, progesterone receptor, HER2, and Ki67.  The authors reported that ROR added significant prognostic information beyond clinical treatment score (CTS) in all subjects (p<0.001) and in all four subgroups: node-negative, node-positive, HER2 negative, and HER2 negative/node negative.  More subjects were scored as high-risk and fewer as intermediate-risk by ROR than by RS.  Relatively similar prognostic information was added by ROR and IHC4 in all subjects, but more by ROR in the HER2 negative/node negative group.  They concluded that ROR provides more prognostic information in endocrine treated subjects with ER positive, node-negative disease than RS, with better differentiation of intermediate and higher-risk groups.

Sestak (2013) described the results of a retrospective non-comparative study involving 940 samples from the ATAC and TransATAC (Translational Substudy of Anastrozole, Tamoxifen, Alone or in Combination) studies.  The study was limited to subjects from the UK who had not previously received chemotherapy, and for whom results from the Oncotype DX, the IHC4, and Prosigna assay (without tumor size) were available.  Median follow-up time was 10 years.  It was reported that the Prosigna derived risk of recurrence (ROR) score was the strongest molecular prognostic factor in the late follow-up period (5-10 years) (χ2=16.29; p<0.001), whereas IHC4 (χ2=7.41) and RS (χ2=5.55) were only weakly prognostic in this period.  Similar results were seen for all subgroups and for all recurrences.  The authors concluded that, except for nodal status and tumor size, none of the IHC4 markers provided statistically significant prognostic information in years 5 to 10.  However, the ROR gave the strong prognostic information for the 5- to 10-year period.  These results are promising, but investigation in a prospective trial is warranted.

In 2015, Sestak published a similar study using combined data from 862 subjects from the ATAC study and 1275 subjects from the Austrian Breast & Colorectal Cancer Study Group (ABCSG-8) study.  All subjects had hormone receptor positive breast cancer treated with endocrine therapy and were free from recurrence 5 years after diagnosis. Mean follow-up was 10 years.  The authors stated that the mean ROR for individuals excluded from this trial due to recurrence within 5 years was significantly higher than the included subjects (ROR=53.57 vs. 20.4, p<0.001).  In the overall population, when compared to the low-risk ROR group, subjects in the high-risk group had a 6.9 fold higher risk of distant recurrence and those in the intermediate-risk ROR group had a 3.3 fold higher risk.  The CTS added more prognostic information for distant recurrence 5 years following diagnosis than the ROR in the overall population (univariate likelihood ratio (LR) χ2=67.94, bivariate LR χ2=35.25).  However, in the node-negative subjects, the ROR score added more prognostic information in both the univariate and bivariate analyses (univariate ROR LR χ2=30.95 and LR χ2=21.48; bivariate CTS LR χ2=17.25 and LR χ2=7.79).  In this subpopulation, subjects in the low-risk ROR group had a 2% risk of distant recurrence by 10 years compared to 9% for the intermediate-risk group and 11.5% for the high-risk group.  In node-positive subjects, the CTS provided the most prognostic information (LR χ2=35.60).  The correlation of CTS and ROR was weak (r=0.36).  Agreement between the two scores was similar for the low-risk groups, but the CTS categorized more subjects into the intermediate-risk group (32.4% vs. 25.2%) and the ROR categorized more subjects into the high-risk group (19.5% vs. 14.3%).  When the classification using ROR was compared to classification using the CTS alone, net reclassification was 7.4%.  When classification using the ROR plus the CTS was compared to the CTS alone, the net reclassification was also 7.4%.

In 2014, Gnant and others described the findings of a retrospective study using banked samples from 1478 post-menopausal subjects who participated in the ABCSG-8 trial.  Both PAM50 intrinsic subtypes and ROR score were calculated for each sample.  The authors reported that in all tested subgroups, ROR score significantly added prognostic information to the clinical predictor (p<0.0001).  PAM50 assigned an intrinsic subtype to all cases, and the Luminal A cohort had a significantly lower ROR at 10 years compared with Luminal B (p<0.0001).  The authors reported significant and clinically relevant discrimination between low- and high-risk groups occurred also within all tested subgroups. 

Gnant (2015) published another study using data from 543 evaluable tissue samples from subjects involved in the TransATAC and ABCSG-8 studies with one to three positive nodes.  Mean follow-up was 9.6 years.  Overall, 97 distant recurrence-free survival events were reported.  In the subjects with one positive node, absolute 10-year risk of distant recurrence was 6.6% in the low-risk ROR group vs. 15.5% in the intermediate group, 25.5% in the high-risk group (p=0.0002 for intermediate and high-risk groups combined), 8.4% in the luminal A subgroup, and 25.3% in the luminal B subgroup (p=0.0001 for the luminal A and luminal B subgroup combined).  The predictive power of CTS with regard to 10-year risk of distant recurrence was significantly improved by the addition of the ROR when stratified by the number of nodes with metastases (p<0.0001, for 1, 2, and 3 nodes).  When subjects in each nodal subgroup were further stratified by ROR risk category, there were no significant differences between the low and intermediate groups.  However, the probability of distant recurrence was significantly increased in the high-risk group vs. the low-risk group (HR=3.56; p=0.0016).

Finally, Filipits and colleagues (2014) describe their findings of 1246 subjects enrolled in the ABCSG-8 study whose banked samples underwent PAM50 ROR testing.  They reported that PAM50 ROR score and ROR-based risk groups provided significant additional prognostic information with respect to late DRFS compared with a combined score of clinical factors alone (ROR score, p<0.001; ROR-based risk groups, p<0.001).  Looking at clinical outcomes between years 5 and 15, the absolute risk of distant recurrence was found to be 2.4% in the low ROR-based risk group, vs. 17.5% in the high ROR-based risk group.  The DRFS differences according to the PAM50 ROR score were observed for both node-positive and node-negative disease.

At this time there is adequate evidence to demonstrate that the Prosigna assay provides sufficient data to assist in the initial prognosis of select individuals with breast cancer

Breast Cancer Index

The Breast Cancer Index (BCI) is a test that combines a two-gene ratio HOXB13:IL17BR (H:I) with a five-gene molecular grade index.  It has been proposed as an aid in the prediction of recurrence of breast cancer in individuals with ER+, lymph node-negative tumors.  A blind, retrospective analysis of data from a prospective RCT of tamoxifen efficacy (Stockholm trial) was published by Jerevall and others (2011).  The Jerevall study involved 588 samples from 314 subjects who were ER+ and received tamoxifen therapy, and 274 subjects who were ER- and did not receive endocrine therapy.  All subjects were considered at low-risk of recurrence, were node-negative and had tumors < 3 cm.  The authors report that in a multivariate Cox linear regression model of tamoxifen-treated subjects, the BCI was a prognostic factor of breast cancer specific death, independent of tumor size, grade, HER2 status, and PR status (p=0.05).  In untreated subjects, the BCI was prognostic for distant recurrence and breast cancer specific death independent of tumor size, grade, HER2 status and PR status (p=0.003).  The authors note that the data for this study was derived from a study conducted from 1976 to 1990 and acknowledge that significant changes in practice patterns have occurred.  They conclude that, “Further studies are warranted to determine whether these findings will extend to current standard of care of ER positive patients receiving 5-10 years of aromatase inhibitors.”

Jankowitz and others (2011) describe the results of a study involving tumor samples from 265 subjects with ER+, lymph node-negative breast cancer who had received adjuvant tamoxifen.  Samples were assessed with the BCI and compared to AOL results.  The authors built a proportional hazard model using known prognostic clinical variables (age, tumor size, tumor grade, and treatment).  When BCI was included in the model, it was highly significant and associated with recurrence risk (p=0.0002), all-cause mortality (p<0.0001), and breast cancer specific mortality (p<0.0001).  A combined multivariate analysis with only BCI and AOL showed that both remained independently and significantly associated with risk of recurrence (p=0.0004 and p=0.0007, respectively), all-cause mortality (p=0.009 and p<0.0001, respectively), and breast cancer specific mortality (p=0.009 and p=0.0001, respectively).  To assess accuracy, the concordance between BCI and AOL were compared to survival times gathered from subject records over a period of 10 years (iAUC).  For time to distant recurrence for all subjects, iAUC values were 0.642 for models with AOL only and 0.717 for models combining AOL and BCI.  For the subjects treated with tamoxifen alone, these probability values increased to an iAUC of 0.671 and 0.750 for models, respectively.  Further analysis indicates that for a time period less than 4 years, the addition of BCI to predictive models provided no significant benefits.  For time periods between 4 and 10 years, the minimum predictive accuracy was 61.4% to 66.2% for AOL only and 71.1% to 75.7% for AOL plus BCI.  For subjects receiving tamoxifen alone, the ranges for this time period were 54.6% to 65.0% for AOL and 73.4% to 80.7% for AOL plus BCI.  The findings of this study are limited by its retrospective nature, as well as having come from a single institution that is a tertiary care center, which may introduce selection bias.  Additionally, limitations in sample availability also may have introduced confounding factors. 

Mathieu and colleagues (2012) reported on the results of a blinded retrospective analysis of 150 subjects previously treated with surgery for breast cancer and treated with chemotherapy and/or tamoxifen.  Ninety-seven percent of subjects had T1-T3 tumors, and 68% were ER+.  Using the BCI assay, 64 subjects (42%) were classified as low-risk, 52 (35%) subjects were classified as intermediate-risk, and 34 subjects (23%) as high-risk.  BCI risk categories were significantly associated with tumor grade (p<0.0001) and ER/PR status (p=0.0013).  BCI was also associated with an 18-fold increased likelihood of pathological complete response between high-risk (n=10/34, 29%) vs. low-risk subjects (n=1/64, 1.6%; p=0.0001).  Within a multivariate analysis which included age, ER/PR, grade, size and HER2 status, BCI remained significantly associated with pathological complete response, both as a continuous score (p=0.0013) and as categorized risk groups with an odds ratio of 34 for high vs. low-risk (p=0.0055).  The authors stratified BCS eligibility BCI, and reported a more than threefold increase in the percentage of subjects undergoing conservative surgery within intermediate-risk or high vs. low-risk groups (p=0.0002).  In multivariate analysis, tumor size and BCI risk group remained significantly associated with BCS with an odds ratio of 5.78 for high vs. low-risk groups (p<0.0001 and p=0.0022, respectively).  The authors acknowledge that the study design may have increased the predictive power of the BCI due to subjects not being selected based on ER or HER2 expression for the indications of neoadjuvant chemotherapy.  Additionally, the retrospective nature of the trial also impacts the generalizability of the findings.

In 2013, Sgroi reported the results of a prospective comparative study investigating the prognostic value of the BCI, Oncotype DX, and an immunohistochemical prognostic model (IHC4).  This study used archival tumor blocks from the TransATAC tissue bank from all post-menopausal individuals with estrogen receptor positive breast cancer from whom the 21-gene recurrence score and IHC4 values had already been derived.  BCI testing was conducted on all samples with sufficient residual RNA using two BCI models, cubic (BCI-C) and linear (BCI-L).  The prognostic ability of BCI for distant recurrence over 10 years was compared with that of results from the Oncotype DX test and the IHC4.  The ability to predict early (0-5 years) and late (5-10 years) distant recurrence was also evaluated.  Testing was successfully done in 665 samples.  No significant difference in risk of distant recurrence over 10 years was noted using the BCI-C test compared to the other tests (p<0.0001).  However, the BCI-L test was a much stronger predictor for overall (0-10 year) distant recurrence compared with BCI-C (p<0.0001).  Compared with BCI-L, the Oncotype DX test was less predictive (HR=1.4, p=0.0002) and IHC4 was similar (HR=1.69, p<0.0001).  In a multivariate analysis, all assays had significant prognostic ability for early distant recurrence (BCI-L: HR=2.77, p<0.0001; Oncotype DX: HR=1.80, p<0.0001; IHC4: HR=2.90, p<0.0001).  However, only BCI-L was significant for late distant recurrence (p=0.0048).  The authors concluded that:

The BCI-L test was the only significant prognostic test for risk of both early and late distant recurrence and identified two risk populations for each timeframe.  It could help to identify patients at high risk for late distant recurrence who might benefit from extended endocrine or other therapy.

This study only included post-menopausal women, limiting the generalizability of these results.  Additionally, the main conclusions are made based upon a secondary linear combination analysis, not on primary analysis.

Zhang and colleagues (2013) reported the results of a study testing the prognostic performance of the BCI in a population of 675 subjects who were enrolled in two separate trials.  The first, referred to as the Stockholm study, enrolled 317 post-menopausal women with node-negative early-stage invasive breast cancer treated with either tamoxifen or no additional treatment.  Subjects were followed for a mean of 17 years in this study.  The second study, referred to as the multi-institutional study is described above (Jankowitz, 2011) and involved tumor samples from 265 subjects with ER+, lymph node-negative breast cancer who had received adjuvant tamoxifen.  Subjects were followed for a mean of 10 years.  Using data from a trial set of 283 subjects from the Stockholm trial data, an optimized BCI was developed for subjects with ER+, lymph node -negative breast cancer.  In this population, 156 subjects were classified as low-risk, 75 as intermediate-risk, and 52 as high-risk.  The 10-year rates of distant recurrence were reported to be 11.2%, 21.0% and 34.4%, respectively.  Data from the Stockholm and multi-institutional study were used to validate the performance of the BCI.  In the Stockholm study population (n=317), early 5-year DMFS was reported to be 98% in the low-risk group, 95.2% in the intermediate-risk group, and 87.8% in the high-risk group (p=0.0063).  For late recurrence (greater than 5 years) 285 subjects were included for analysis.  In this population, 10-year DMFS was reported to be 97.2%, 92.8% and 89.9% in the low, intermediate, and high-risk classification groups, respectively (p=0.0152).  Using data from the multi-institutional study (n=358), 5-year DMFS was 95.5%, 92.3% and 75.5% % in the low, intermediate, and high-risk classification groups, respectively (p<0.0001), and 10-year DMFS was 97.5%, 83.1% and 85.0% in the low, intermediate, and high-risk classification groups, respectively (p=0.0002).

At this time there is adequate evidence to demonstrate that the BCI assay provides sufficient data to assist in the prognosis of select individuals with breast cancer after 5 years of treatment to determine whether or not additional endocrine is warranted.

BreastPRS

D’Alfonso and others (2013) described a validation study using samples of 246 subjects with invasive breast carcinoma and known Oncotype DX results.  They applied the BreastPRS test and a 120-gene Oncotype DX approximation algorithm to a series of untreated, node-negative, estrogen receptor positive (ER+) subjects from previously published studies with known clinical outcomes.  Correlation of recurrence score and risk group between Oncotype DX and BreastPRS was statistically significant (p<0.0001).  Out of 260 subjects, 59 (23%) from four previously published studies were classified as intermediate-risk when the 120-gene Oncotype DX approximation algorithm was applied.  The BreastPRS test reclassified the 59 subjects into binary risk groups (high vs. low-risk), with 23 (39%) subjects classified as low-risk, and 36 (61%) as high-risk.  At 10-years from diagnosis, the low-risk group had a 90% recurrence-free survival (RFS) rate compared to 60% for the high-risk group.  The authors concluded that “BreastPRS recurrence score is comparable with Oncotype DX and can reclassify Oncotype DX intermediate risk subjects into two groups with significant differences in RFS. Further studies are needed to validate these findings.”  A limitation of this study is the use of the “120-gene Oncotype DX approximation algorithm” which, was developed by the authors, instead of the actual Oncotype DX test.  The methodology used undermines the comparison, since the results of the algorithm may or may not reflect the actual Oncotype DX results.

EndoPredict

The EndoPredict assay is based on the quantification of eight cancer-related genes and three normalization genes resulting in an EndoPredict Risk Score (EP).  When the EP is combined with the clinical risk factors of nodal status and tumor size, an EndoPredict Clinical Risk Score (EPclin) is derived.  The results of this test stratify individuals into low and high-risk categories for distant metastases at 5 and 10 years. 

In 2013, Filipits and colleagues reported on a retrospective validation study using 964 tumor samples from ER+, HER2- subjects treated with tamoxifen only, who were enrolled in the ABCSG-6 and ABCSG-8 studies.  The results of a multivariate analysis indicated that EP was an independent predictor of distant recurrence in both ABCSG-6 and ABCSG-8 cohorts.  At 10 years, the distant recurrence rates for the low-EP and high-EP groups were 8% vs. 22% in the ABCSG-6 cohort and 6% and 15% in the ABCSG-8 cohort (p<0.0010 for both).  Similar findings were reported for EPclin, with the distant recurrence rates for the low-EP and high-EP groups at 4% vs. 28% in the ABCSG-6 cohort and 4% vs 22% in the ABCSG-8 cohort (p<0.0001 for both).  The concordance index, (c-index) was calculated for the addition of the EP to both clinicopathologic markers as well as Adjuvant! Online score.  For the clinicopathologic markers, the addition of EP resulted in a significantly improved prognostic power (c-index=0.727 in the ABCSG-6 cohort and 0.728 in the ABCSG-8 cohort).  Similar results were reported for the addition of EP to the Adjuvant! Online score (c-index=0.785 in the ABCSG-6 cohort and 0.733 in the ABCSG-8 cohort).  The EPclin had a higher c-index than any combination of EP and clinicopathologic variables (c-index=0.788 in the ABCSG-6 cohort and 0.732 in the ABCSG-8 cohort).

The same group published additional data from 1702 subjects from the ABCSG-6 and ABCSG-8 studies (Dubsky, 2013a).  In this study, all subjects were assigned to a risk category based on cut-off values from the German S3, NCCN, and St. Gallen treatment recommendations.  Low-risk categorization was assigned to 15% of subjects with German S3, 6% with the NCCN, 19% with the St. Gallen recommendations and 63% with EPclin.  Absolute freedom of distant recurrence (AFDR) at 10 years was reported to be 94.7%, 94.5%, 96.6% and 95.3%, respectively.  Using EPclin, AFDR was reported as 95.3%.  The authors compared the DMFS for the low-risk group vs. the intermediate and high-risk groups for each guideline.  For the German S3, DMFS was significantly better than the other two risk groups (p=0.014, HR-2.2 and absolute risk reduction [ARR] of 7.9%).  No significant differences were reported for NCCN (p=0.12, HR=2.16, and ARR=6.9%).  St. Gallen resulted in a significant association with decreased 10-year DMFS (p<0.001, HR=2.78, and ARR=11.2%).  Use of the EPclin resulted in the best separation between the low, and high-risk groups (p<0.001, HR=5.11 and ARR=18.7%). 

A third study from this group was published later in the same year (Dubsky, 2013b), using the same dataset.  Overall, 49% of subjects (n=832) were classified as low-risk by EP.  Analysis indicated that this cohort had significantly improved clinical outcomes in the first 5 years (p<0.001), and beyond 5 years (p=0.002).  This group also had an absolute freedom of distant recurrence of 96.29% between 5 and 10 years.  As with their earlier study, c-indices were reported.  The addition of clinical parameters to the EP value (EPclin) increased the c-index from 0.644 to 0.716 (p<0.001).  The addition of the EP to the Adjuvant! Online score resulted in the c-index improving from 0.674 to 0.756 (no p-value provided).  At 10 years, the absolute freedom from distant recurrence in the EPclin low-risk and EPclin high-risk groups was reported to be 98.2% and 87.69%, respectively.

Martin and others (2014) reported the results of another validation study involving 555 ER+, HER2-, lymph node+ subjects enrolled in the chemotherapy-treated arm of the GEICAM 9906 study.  In this study, the estimated DMFS at 10 years was 93% for the low-risk EP group and 70% in the high-risk EP group.  The ARR was calculated at 23% (HR=4.8, p<0.0001).  As with the previously described studies, c-index values were reported.  The authors reported that the c-index for clinicopathologic parameters alone was 0.065 and increased to 0.67 when EP was added (p<0.0018).  For EPclin, 10-year estimates of DMFS were 100% in the low-risk group and 72% in the high-risk group with an ARR of 28% (p<0.001).  When stratified for menopausal status, the data indicated that EP was prognostic for distant metastases in pre-menopausal subjects (HR=6.7, p<0.0001) as well as post-menopausal subjects (HR=3.3, p<0.0069).  The results for EPclin were similar (p=0.0006 and 0.0023, respectively).

In 2015, Fitzal reported the results of a retrospective cohort study that used tumor blocks from 1324 subjects who participated in the ABCSG 8 trial to predict local recurrence-free survival (LRFS) following surgery for breast cancer in post-menopausal women.  Median follow-up was 72.3 months and the cumulative incidence of local recurrence was 2.6% (0.4% per year).  The risk of local recurrence over a 10 year period among subjects with lesions identified as high-risk by the EndoPredict test (n=683) was significantly higher (10 year LRFS=91%) when compared to subjects identified with low-risk lesions (n=641; 10 year LRFS=97.5%) (HR=1.31; p<0.005).  The groups that received breast conservation surgery and mastectomy had similar local recurrence rates (p=0.879).  RT after breast conservation surgery significantly improved LRFS in the cohorts predicted by EndoPredict to be low-risk for LR (p<0.005).  The authors concluded that among post-menopausal, low-risk individuals, use of the EndoPredict test does not appear to be useful for tailoring local therapy.  

In 2016, Buus and colleagues published the results of a retrospective study involving 928 samples from the TransATAC trial for women with localized primary ER+, HER2- breast cancer, who had not received chemotherapy and who had sufficient material for EndoPredict (EP) and EndoPredict Clinical analysis.  TransATAC also served as a validation study for Oncotype DX and allowed comparisons between EP and the Oncotype Risk Score (RS).  Comparisons were made for distant recurrences between 0-5 years and 5-10 years.  Distant recurrence occurred in 128 subjects within 10 years (59/680 node-negative and 69/248 node-positive subjects).  Both EP and EPclin were highly prognostic across 10 years, with EPclin being significantly more prognostic than EP alone (LRχ2: EP=49.3; LRχ2: EPclin=139.3).  Both were more prognostic than the Oncotype DX RS over 10 years in the total subject pool (LRχ2: RS=29.1).  Both EP and EPclin provided significant prognostic value vs. RS with later recurrences, 5-10 years, for all subjects (LRχ2: EPclin=59.3, p<0.001; LRχ2: EP=23.6, p<0.001; LRχ2: RS=5.6 p=0.02) and for node-positive subjects (LRχ2: EPclin=48.3, p<0.001; LRχ2: EP=14.5, p<0.001; LRχ2: RS=8.0, p=0.005).  EPclin was particularly good at stratifying node-positive subjects with an absolute separation at 10 years for distant recurrence of 31.9% vs. 14.1% in node-negative subjects.  The HR between high- and low-risk groups was marginally greater for EP (HR=2.98, p<0.001) than for RS (HR=2.73, p<0.001) and substantially greater than for EPclin (HR=5.99, p<0.001).  In the majority of cases, EPclin and RS were in agreement (K=0.40), but 117 cases (12.6%) were classified as EPclin high-risk and RS non-low-risk.  Additionally, 144 cases (15.5%) were EPclin high-risk and RS low-risk.  The authors concluded that EP and EPclin were highly prognostic for distant recurrence in endocrine-treated subject with ER+, HER2- breast cancer, and that EPclin provided more prognostic information than the Oncotype DX RS.

At this time, there is adequate evidence to demonstrate that both the EndoPredict and the EPclin assays provide sufficient data to assist in the prognosis of select individuals with breast cancer, both as an initial test prior to treatment and after 5 years of treatment to determine whether or not additional endocrine therapy is warranted.

Systematic Reviews and Guidelines

In March 2008, Marchionni and colleagues published a systematic review of the data addressing the use of gene expression profiling tests in breast cancer.  In that document the authors conclude:

Although these tests show great promise to improve predictions of prognosis and treatment benefit for women with early stage breast cancer, more must be learned about the extent of that improvement, in whom it is most improved, and how the tests are best incorporated into decision making about current breast cancer treatment.

A study by Prat and colleagues published in 2012 addressed the level of concordance between gene expression breast cancer prediction tests.  The study describes the combination of four separate microarray datasets totaling 594 ER+ subjects who received local therapy and tamoxifen which were then evaluated using the Oncotype DX, MammaPrint, Rotterdam 76-gene signature, and the PAM50 assay tests to predict relapse in the sample population.  The authors report that all tests were prognostic in subjects with ER+, node-negative tumors, whereas most were prognostic in ER+, node-positive disease.  Among the signatures evaluated, PAM50, OncotypeDX, and MammaPrint were consistently found to be independent predictors of relapse.  A combination of all signatures significantly increased the performance prediction.  Importantly, low-risk tumors (> 90% distant relapse-free survival at 8.5 years) were identified by the majority of signatures only within node-negative disease, and these tumors were mostly Luminal A (78%-100%).  The authors comment that while these tests provide statistically independent results from each other, they should not be considered interchangeable.  This is highlighted by their finding that each test identifies a separate subset of node-negative subjects who might be suitable candidates for chemotherapy.

Recommendations from the 14th St. Gallen International Breast Cancer Conference (2015), in reviewing therapies for the management of early breast cancer, noted the increasing evidence for the prognostic value of multiparametric molecular markers, stating, “Oncotype DX®, MammaPrint®, PAM-50 ROR® score, EndoPredict®, and the Breast Cancer Index® were all considered usefully prognostic for years 1-5.”  Beyond 5 years, only the PAM50 ROR score was agreed to be clearly prognostic.  The panel further states, “Only Oncotype DX® commanded a majority in favor of its value in predicting the usefulness of chemotherapy.”

The ASCO recommendations (Harris, 2016) addressing the appropriate use of breast tumor biomarker assay results to guide decisions on adjuvant systemic therapy in early-stage invasive breast cancer, notes that the panel found sufficient evidence of clinical utility for the biomarker assays Oncotype DX, EndoPredict, PAM50 ROR, and Breast Cancer Index. Stating further that, “No biomarker except for estrogen receptor, progesterone receptor, and human epidermal growth factor receptor 2 was found to guide choices of specific treatment regimens.”  The panel supported their statement on the clinical utility of assays in guiding decisions on adjuvant systemic chemotherapy by rating the type of evidence as “evidence based”, the quality at “high” and strength of recommendation as “strong” for Oncotype DX and the PAM50 ROR; and for the EndoPredict and Breast Cancer Index assays as type of evidence “evidence based”, quality “intermediate” and strength of recommendation “moderate.”

It should be noted that in making these statements, ASCO utilized the following definition of clinical utility (Harris, 2016, supplemental material): “The use of test results to guide clinical decisions improved measurable outcomes of patient management, compared to decisions independent of test results.”  With regard to their recommendations for the PAM50 test, this standard is not met.

In 2017, ASCO published an update to the 2016 recommendations, specifically addressing the MammaPrint assay (Krop, 2017).  Their 2017 strong recommendations, based primarily on the results of the MINDACT study published by Cardoso (2016), support the use of MammaPrint for patients with high clinical risk per the MINDACT categorization, but do not support use in patients with low clinical risk.

The NCCN Practice Guidelines in Oncology: Breast Cancer (v.2.2017) made the following statement in a footnote addressing the use of 21-GENE RT-PRC, in the most recent update, “Other prognostic multigene assays may be considered to help assess risk of recurrence but have not been validated to predict response to chemotherapy.” 

At this time, the aggregate data do not support the use of the MammaPrint, Breast Cancer Gene Expression Ratio test (H:I Ratio), the Rotterdam 76-gene assay, the Insight DX Breast Cancer Profile, the PAM50 Breast Cancer Intrinsic Classifier, the Breast Cancer Index, or the BreastPRS.  These tests are not currently recommended by the NCCN guidelines. 

General Comments

The NCCN guidelines (V.2.2017) state that, “There are limited data to make chemotherapy recommendations for those over 70 years old.  Treatment should be individualized with consideration of comorbid conditions.”  Thus, the use of the gene expression profiling in this population may not be warranted.

There are many other breast cancer profiling tests available on the market for which there are no available studies published in the peer-reviewed published literature, including but not limited to the following tests: BreastNext, BreastOncPX, MammaTyper, NexCourse Breast IHC4, NuvoSelect eRx 200-Gene Assay, SYMPHONY Genomic Breast Cancer Profile, TargetPrint, TheraPrint, and THEROS Breast Cancer Index.  For these tests, the safety and efficacy of their use in the care of individuals with breast cancer is unknown and unproven.

Background/Overview

Clinical evidence has demonstrated that among individuals with breast cancer there is a continuum of disease recurrence risk, based on many factors including age, the presence of various hormone receptors in tumor samples, tumor size, whether or not the cancer has spread outside the breast, and others.  Clinicians have, for practical use, divided this continuum up into three risk categories: (1) low-risk, (2) intermediate-risk, and (3) high-risk.  These risk categories have been used as a method of helping to determine what treatment methods to use for specific individuals.  In individuals deemed at high-risk for disease recurrence, the medical evidence has shown that the use of chemotherapy in addition to other treatment may provide a significant survival benefit.  In low-risk individuals, the data has shown that chemotherapy in addition to other treatment does not provide any significant benefits.  However, the available information regarding whether or not intermediate-risk individuals benefit from chemotherapy is unclear.  Traditionally, treating clinicians have to balance each individual’s risk of disease recurrence with the risks of chemotherapy, which include hair loss, nausea, vomiting, weakness, infection, and others.

Recently a new type of test, the gene expression profiling assay, has been developed to help clinicians determine which populations of intermediate-risk individuals would benefit from chemotherapy.  Gene expression profiling assays measure the presence of a variety of genes which have been associated with the recurrence of breast cancer.  Using these tests, in conjunction with other traditional risk assessment methods, clinicians may be able to more accurately determine which intermediate-risk individuals would benefit from chemotherapy, and which individuals would not.  In this way individuals most likely to benefit from chemotherapy are identified and receive needed care, and those individuals who would not benefit are spared the unnecessary treatment and risks associated with chemotherapy without adversely affecting disease-free and overall survival outcomes.

Definitions

Ductal carcinoma in situ (DCIS): An early, non-invasive form of breast cancer, where abnormal cells grow inside a milk duct in the breast.

Estrogen receptor status: A laboratory finding related to the presence or absence of cellular receptors for the hormone estrogen.

HER2 receptor status: A laboratory finding related to the presence or absence of cellular receptors for HER2/neu (also known as ErbB-2, ERBB2) protein family. HER2 is notable for its role in the pathogenesis of breast cancer and as a target of treatment. There are two different methods by which HER2 receptor status can be discovered; the first is immunohistochemistry (IHC) and the other is in situ hybridization (ISH). According to the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) (Wolff, 2013) and NCCN (V.2.2017) recommendations for HER2 testing for breast cancer, scoring of these tests is as follows:

Note 1:  When HER2 testing by validated IHC test is equivocal: “Must reflex test with ISH (if same specimen), or order new test with IHC (if new specimen is available).”
Note 2:   When HER2 testing validated by a single-probe ISH assay is equivocal: “Must reflex test with dual-probe ISH or with IHC (if same specimen), or order new test with SIH or IHC (if new specimen available).”
Note 3:   When HER2 testing done by a validated dual-probe ISH assay is equivocal:  “Must reflex test with IHC (if same specimen), test with alternative ISH chromosome 17 probe, or order a new test with ISH or IHC (if new specimen is available).”

Note 1:  For equivocal results on IHC tests, the NCCN states, “Must reflex test with ISH (if
same specimen) or order new test with IHC or ISH (if new specimen available).

*NCCN Clinical Practice Guidelines in Oncology: Breast Cancer (V.2.2017). April 6, 2017. page BINV-A.

Histology: A method of categorizing tissues by evaluating cells and tissues at the cellular level with microscopic examination; the following are several histological categories of breast cancer:

Progesterone receptor status: A laboratory finding related to the presence or absence of cellular receptors for the hormone progesterone.

TNM (tumor size, nodal involvement, and metastasis) classification: A classification system used for characterizing the size and location of breast cancer. The system is described below:
(Note: A prefix of “p” indicates by pathology report)

Primary Tumor:

TX:     Primary tumor cannot be assessed
T0:      No evidence of primary tumor
Tis:     Carcinoma in situ; intraductal carcinoma, lobular carcinoma in situ, or Paget's disease of the nipple with no associated tumor.  Note: Paget's disease associated with a tumor is classified according to the size of the tumor.
T1:      Tumor 2.0 cm or less in greatest dimension
T1mic:  Microinvasion 0.1 cm or less in greatest dimension
T1a:      Tumor more than 0.1 but not more than 0.5 cm in greatest dimension
T1b:      Tumor more than 0.5 cm but not more than 1.0 cm in greatest dimension
T1c:    Tumor more than 1.0 cm but not more than 2.0 cm in greatest dimension
T2:      Tumor more than 2.0 cm but not more than 5.0 cm in greatest dimension
T3:      Tumor more than 5.0 cm in greatest dimension
T4:      Tumor of any size with direct extension to (a) chest wall or (b) skin, only as described below. Note: Chest wall includes ribs, intercostal muscles, and serratus anterior muscle but not pectoral muscle.
T4a:    Extension to chest wall
T4b:    Edema (including peau d'orange) or ulceration of the skin of the breast or satellite skin nodules confined to the same breast
T4c:    Both of the above (T4a and T4b)
T4d:    Inflammatory carcinoma*

Regional lymph nodes (N):

NX:     Regional lymph nodes cannot be assessed (e.g., previously removed)
N0:      No regional lymph node metastasis
N1:      Metastasis to movable ipsilateral axillary lymph node(s)
N2:      Metastasis to ipsilateral axillary lymph node(s) fixed to each other or to other structures
N3:      Metastasis to ipsilateral internal mammary lymph node(s)

Pathologic classification (pN):

pNX:   Regional lymph nodes cannot be assessed (not removed for pathologic study or previously removed)
pN0:    No regional lymph node metastasis
pN1:    Metastasis to movable ipsilateral axillary lymph node(s)
pNmi:     Micrometastasis (larger than 0.2mm but no larger than 2.0mm)
pN1a:      Only micrometastasis (none larger than 0.2 cm)
pN1b:      Metastasis to lymph node(s), any larger than 0.2 cm
pN1bi:     Metastasis in 1 to 3 lymph nodes, any more than 0.2 cm and all less than 2.0 cm in greatest dimension
pN1bii:    Metastasis to 4 or more lymph nodes, any more than 0.2 cm and all less than 2.0 cm in greatest dimension
pN1biii:   Extension of tumor beyond the capsule of a lymph node metastasis less than 2.0 cm in greatest dimension
pN1biv:   Metastasis to a lymph node 2.0 cm or more in greatest dimension pN2: Metastasis to ipsilateral axillary lymph node(s) fixed to each other or to other structures
pN3:    Metastasis to ipsilateral internal mammary lymph node(s)

Distant metastasis (M):

MX:    Presence of distant metastasis cannot be assessed
M0:     No distant metastasis
M1:     Distant metastasis present (includes metastasis to ipsilateral supraclavicular lymph nodes)

Coding

The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member’s contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.

When services may be Medically Necessary when criteria are met for Breast Cancer Index, EndoPredict, Oncotype DX, or Prosigna Breast Cancer Prognostic Gene Signature Assay:

CPT

 

81519

Oncology (breast), mRNA, gene expression profiling by real-time RT-PCR of 21 genes, utilizing formalin-fixed paraffin embedded tissue, algorithm reported as recurrence score
Oncotype DX (Genomic Health)

0008M

Oncology (breast), mRNA analysis of 58 genes using hybrid capture, on formalin-fixed paraffin-embedded (FFPE) tissue, prognostic algorithm reported as a risk score
Prosigna Breast Cancer Prognostic Gene Signature Assay (Nanostring Technologies) [code to be deleted 12/31/17]

81520

Oncology (breast), mRNA gene expression profiling by hybrid capture of 58 genes (50 content and 8 housekeeping), utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as a recurrence risk score

Prosigna Breast Cancer Assay, NanoString Technologies, Inc. [code effective 01/01/2018]

81599

Unlisted multianalyte assay with algorithmic analysis [when specified as EndoPredict or the Breast Cancer Index]

84999

Unlisted chemistry procedure [when specified as EndoPredict or the Breast Cancer Index]

 

 

HCPCS

 

S3854

Gene expression profiling panel for use in the management of breast cancer treatment [when specified as EndoPredict or the Breast Cancer Index]

 

 

ICD-10 Diagnosis

 

C50.011-C50.929

Malignant neoplasm of breast

C79.81

Secondary malignant neoplasm of breast

D05.00-D05.02

Lobular carcinoma in situ of breast

D05.80-D05.92

Other and unspecified type of carcinoma in situ of breast

Z17.0

Estrogen receptor positive status [ER+]

Z85.3

Personal history of malignant neoplasm of breast

When services are Not Medically Necessary:
For the procedure and diagnosis codes listed above when criteria are not met.
 

When services are Investigational and Not Medically Necessary:
For the procedure codes listed above for all other diagnoses or for situations indicated in the Position Statement section as investigational and not medically necessary.

CPT

 

81521

Oncology (breast), mRNA, microarray gene expression profiling of 70 content genes and 465 housekeeping genes, utilizing fresh frozen or formalin-fixed paraffin-embedded tissue, algorithm reported as index related to risk of distant metastasis

MammaPrint, Agendia, Inc. [code effective 01/01/2018]

 

 

ICD-10 Diagnosis

 

 

All diagnoses

When services are also Investigational and Not Medically Necessary:
For any other type of gene expression testing for breast cancer, or when the code describes a procedure indicated in the Position Statement section as investigational and not medically necessary.

CPT

 

81599

Unlisted multianalyte assay with algorithmic analysis [when specified as a breast cancer gene expression profile other than Oncotype DX, Prosigna, EndoPredict or the Breast Cancer Index]

84999

Unlisted chemistry procedure [when specified as a breast cancer gene expression profile other than Oncotype DX, Prosigna, EndoPredict or the Breast Cancer Index]

 

 

HCPCS

 

S3854

Gene expression profiling panel for use in the management of breast cancer treatment [when specified as a breast cancer gene expression profile other than Oncotype DX, Prosigna, EndoPredict or the Breast Cancer Index]

 

 

ICD-10 Diagnosis

 

 

All diagnoses

References

Peer Reviewed Publications:

  1. Ahr A, Holtrich U, Solbach C, et al. Molecular classification of breast cancer patients by gene expression profiling. J Pathol. 2001; 195(3):312-320.
  2. Ahr A, Karn T, Solbach C, et al. Identification of high risk breast-cancer patients by gene expression profiling.  Lancet. 2002; 359(9301):131-132.
  3. Albain KS, Barlow WE, Shak S, et al. Prognostic and predictive value of the 21-gene recurrence score assay in postmenopausal women with node-positive, oestrogen-receptor-positive breast cancer on chemotherapy: a retrospective analysis of a randomised trial. Lancet Oncol. 2010; 11(1):55-65.
  4. Alvarado M, Carter DL, Guenther JM, et al. The impact of genomic testing on the recommendation for radiation therapy in patients with ductal carcinoma in situ: a prospective clinical utility assessment of the 12-gene DCIS score™ result. J Surg Oncol. 2015; 111(8):935-940.
  5. Ayers M, Symmans WF, Stec J, et al. Gene expression profiles predict complete pathological response to neoadjuvant paclitaxel and fluorouracil, doxorubicin, and cyclophosphamide chemotherapy in breast cancer. J Clin Oncol. 2004; 22(12):2284-2293.
  6. Bartlett JM, Bayani J, Marshall A, et al. Comparing breast cancer multiparameter tests in the OPTIMA prelim trial: no test is more equal than the others. J Natl Cancer Inst. 2016; 108(9).
  7. Bartlett JM, Bloom KJ, Piper T, et al. Mammostrat as an immunohistochemical multigene assay for prediction of early relapse risk in the tamoxifen versus exemestane adjuvant multicenter trial pathology study. J Clin Oncol. 2012; 30(36):4477-4484.
  8. Bueno-de-Mesquita JM, Linn SC, Keijzer R, et al. Validation of 70-gene prognosis signature in node-negative breast cancer. Breast Cancer Res Treat. 2009; 117(3):483-495.
  9. Bueno-de-Mesquita JM, van Harten WH, Retel VP, et al. Use of 70-gene signature to predict prognosis of patients with node-negative breast cancer: a prospective community-based feasibility study (RASTER). Lancet Oncol. 2007; 8(12):1079-1087.
  10. Buus R, Sestak I, Kronenwett R, et al. Comparison of EndoPredict and EPclin with Oncotype DX Recurrence Score for prediction of risk of distant recurrence after endocrine therapy. J Natl Cancer Inst. 2016; 108(11).
  11. Buyse M, Loi S, van't Veer L, et al.; TRANSBIG Consortium. Validation and clinical utility of a 70-gene prognostic signature for women with node-negative breast cancer. J Natl Cancer Inst. 2006; 98(17):1183-1192.
  12. Cardoso F, van't Veer LJ, Bogaerts J, et al.; MINDACT Investigators. 70-gene signature as an aid to treatment decisions in early-stage breast cancer. N Engl J Med. 2016; 375(8):717-729.
  13. Chang JC, Makris A, Gutierrez MC, et al. Gene expression patterns in formalin-fixed, paraffin-embedded core biopsies predict docetaxel chemosensitivity in breast cancer patients. Breast Cancer Res Treat. 2008; 108(2):233-240.
  14. Chang JC, Wooten EC, Tsimelzon A, et al. Gene expression profiling for the prediction of therapeutic response to docetaxel in patients with breast cancer. Lancet. 2003; 362(9381):362-369.
  15. Cobleigh MA, Tabesh B, Bitterman P, et al. Tumor gene expression and prognosis in breast cancer patients with 10 or more positive nodes. Clin Cancer Res. 2005; 11(24 Pt 1):8623-8631.
  16. Dabbs DJ, Klein ME, Mohsin SK, et al. High false-negative rate of HER2 quantitative reverse transcription polymerase chain reaction of the Oncotype DX test: an independent quality assurance study. J Clin Oncol. 2011; 29(3):4279-4285.
  17. D'Alfonso TM, van Laar RK, Vahdat LT, et al. BreastPRS is a gene expression assay that stratifies intermediate-risk Oncotype DX patients into high- or low-risk for disease recurrence. Breast Cancer Res Treat. 2013; 139(3):705-715.
  18. de Boer RH, Baker C, Speakman D, et al. The impact of a genomic assay (Oncotype DX) on adjuvant treatment recommendations in early breast cancer. Med J Aust. 2013; 199(3):205-208.
  19. Desmedt C, Piette F, Loi S, et al.; TRANSBIG Consortium. Strong time dependence of the 76-gene prognostic signature for node-negative breast cancer patients in the TRANSBIG multicenter independent validation series. Clin Cancer Res. 2007; 13(11):3207-3214.
  20. Dowsett M, Cuzick J, Wale C, 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.
  21. Dowsett M, Sestak I, Lopez-Knowles E, et al. Comparison of PAM50 risk of recurrence score with oncotype DX and IHC4 for predicting risk of distant recurrence after endocrine therapy. J Clin Oncol. 2013; 31(22):2783-2790.
  22. Drukker CA, Bueno-de-Mesquita JM, Retèl VP, et al. A prospective evaluation of a breast cancer prognosis signature in the observational RASTER study. Int J Cancer. 2013; 133(4):929-936.
  23. Drukker CA, Nijenhuis MV, Bueno-de-Mesquita JM, et al. Optimized outcome prediction in breast cancer by combining the 70-gene signature with clinical risk prediction algorithms. Breast Cancer Res Treat. 2014b; 145(3):697-705.
  24. Drukker CA, van Tinteren H, Schmidt MK, et al. Long-term impact of the 70-gene signature on breast cancer outcome. Breast Cancer Res Treat. 2014a; 143(3):587-592.
  25. Dubsky P, Brase JC, Jakesz R, et al.; Austrian Breast and Colorectal Cancer Study Group (ABCSG). The EndoPredict score provides prognostic information on late distant metastases in ER+/HER2- breast cancer patients. Br J Cancer. 2013a; 109(12):2959-2964.
  26. Dubsky P, Filipits M, Jakesz R, et al.; Austrian Breast and Colorectal Cancer Study Group (ABCSG). EndoPredict improves the prognostic classification derived from common clinical guidelines in ER-positive, HER2-negative early breast cancer. Ann Oncol. 2013b; 24(3):640-617.
  27. Eiermann W, Rezai M, Kümmel S, et al. The 21-gene recurrence score assay impacts adjuvant therapy recommendations for ER-positive, node-negative and node-positive early breast cancer resulting in a risk-adapted change in chemotherapy use. Ann Oncol. 2013; 24(3):618-624.
  28. Esserman LJ, Yau C, Thompson CK, et al. Use of molecular tools to identify patients with indolent breast cancers with ultralow risk over 2 decades. JAMA Oncol. 2017 Jun 29. [Epub ahead of print]
  29. Filipits M, Nielsen TO, Rudas M, et al.; Austrian Breast and Colorectal Cancer Study Group. The PAM50 risk-of-recurrence score predicts risk for late distant recurrence after endocrine therapy in postmenopausal women with endocrine-responsive early breast cancer. Clin Cancer Res. 2014; 20(5):1298-1305.
  30. Filipits M, Rudas M, Jakesz R, et al.; EP Investigators. A new molecular predictor of distant recurrence in ER-positive, HER2-negative breast cancer adds independent information to conventional clinical risk factors. Clin Cancer Res. 2011; 17(18):6012-6020.
  31. Fisher B, Costantino JP, Redmond CK, et al. Endometrial cancer in tamoxifen-treated breast cancer patients: findings from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-14. J Natl Cancer Inst. 1994; 86(7):527-537.
  32. Fisher B, Dignam J, Wolmark N, et al. Tamoxifen and chemotherapy for lymph node-negative, estrogen receptor-positive breast cancer. J Natl Cancer Inst. 1997; 89(22):1673-1682.
  33. Fitzal F, Filipits M, Rudas M, et al.The genomic expression test EndoPredict is a prognostic tool for identifying risk of local recurrence in postmenopausal endocrine receptor-positive, her2neu-negative breast cancer patients randomised within the prospective ABCSG 8 trial. Br J Cancer. 2015; 112(8):1405-1410.
  34. Foekens JA, Atkins D, Zhang Y, et al. Multicenter validation of a gene expression-based prognostic signature in lymph node-negative primary breast cancer. J Clin Oncol. 2006; 24(11):1665-1671.
  35. Gianni L, Zambetti M, Clark K, et al. Gene expression profiles of paraffin-embedded core biopsy tissue predict response to chemotherapy in women with locally advanced breast cancer. J Clin Oncol. 2005; 23(29):7265-7277.
  36. Glück S, de Snoo F, Peeters J, et al. Molecular subtyping of early-stage breast cancer identifies a group of patients who do not benefit from neoadjuvant chemotherapy. Breast Cancer Res Treat. 2013; 139(3):759-767.
  37. Gluz O, Nitz UA, Christgen M, et al. West German Study Group Phase III PlanB Trial: First prospective outcome data for the 21-gene recurrence score assay and concordance of prognostic markers by central and local pathology assessment. J Clin Oncol. 2016; 34(20):2341-2349.
  38. Gnant M, Filipits M, Greil R, et al.; Austrian Breast and Colorectal Cancer Study Group. Predicting distant recurrence in receptor-positive breast cancer patients with limited clinicopathological risk: using the PAM50 Risk of Recurrence score in 1478 postmenopausal patients of the ABCSG-8 trial treated with adjuvant endocrine therapy alone. Ann Oncol. 2014; 25(2):339-345.
  39. Gnant M, Sestak I, Filipits M, et al. Identifying clinically relevant prognostic subgroups of postmenopausal women with node-positive hormone receptor-positive early-stage breast cancer treated with endocrine therapy: a combined analysis of ABCSG-8 and ATAC using the PAM50 risk of recurrence score and intrinsic subtype. Ann Oncol. 2015; 26(8):1685-1691.
  40. Goetz MP, Suman VJ, Ingle JN, et al. A two-gene expression ratio of homeobox 13 and interleukin-17B receptor for prediction of recurrence and survival in women receiving adjuvant tamoxifen. Clin Cancer Res. 2006; 12(7 Pt 1):2080-2087.
  41. Goldstein LJ, Gray R, Badve S, 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):4063-4071.
  42. Iwao-Koizumi K, Matoba R, Ueno N, et al. Prediction of docetaxel response in human breast cancer by gene expression profiling. J Clin Oncol. 2005; 23(3):422-431. 
  43. Jankowitz RC, Cooper K, Erlander MG, et al. Prognostic utility of the breast cancer index and comparison to Adjuvant! Online in a clinical case series of early breast cancer. Breast Cancer Res. 2011; 13(5):R98.
  44. Jerevall PL, Brommesson S, Strand C, et al. Exploring the two-gene ratio in breast cancer-independent roles for HOXB13 and IL17BR in prediction of clinical outcome. Breast Cancer Res Treat. 2008; 107(2):225-234.
  45. Jerevall PL, Ma XJ, Li H, et al. Prognostic utility of HOXB13:IL17BR and molecular grade index in early-stage breast cancer patients from the Stockholm trial. Br J Cancer. 2011; 104(11):1762-1769.
  46. Kaklamani VG, Gradishar WJ. Gene expression in breast cancer. Curr Treat Options Oncol. 2006; 7(2):123-128.
  47. Kelly CM, Bernard PS, Krishnamurthy S, et al. Agreement in risk prediction between the 21-gene recurrence score assay (Oncotype DX®) and the PAM50 breast cancer intrinsic Classifier™ in early-stage estrogen receptor-positive breast cancer. Oncologist. 2012; 17(4):492-498.
  48. Knauer M, Mook S, Rutgers EJ, et al. The predictive value of the 70-gene signature for adjuvant chemotherapy in early breast cancer. Breast Cancer Res Treat. 2010; 120(3):655-661.
  49. Kok M, Koornstra RH, Mook S, et al. Additional value of the 70-gene signature and levels of ER and PR for the prediction of outcome in tamoxifen-treated ER-positive breast cancer. Breast. 2012; 21(6):769-778.
  50. Linke SP, Bremer TM, Herold CD, et al. A multimarker model to predict outcome in tamoxifen-treated breast cancer patients. Clin Cancer Res. 2006; 12(4):1175-1183.
  51. Ma XJ, Salunga R, Dahiya S, et al. A five-gene molecular grade index and HOXB13:IL17BR are complementary prognostic factors in early stage breast cancer. Clin Cancer Res. 2008; 14(9):2601-2608.
  52. Mamounas EP, Tang G, Fisher B, et al. Association between the 21-gene recurrence score assay and risk of locoregional recurrence in node-negative, estrogen receptor-positive breast cancer: results from NSABP B-14 and NSABP B-20. J Clin Oncol. 2010; 28(10):1677-1683.
  53. Marchionni L, Wilson RF, Wolff AC, et al. Systematic review: gene expression profiling assays in early-stage breast cancer. Ann Intern Med. 2008; 148(5):358-369. 
  54. Mathieu MC, Mazouni C, Kesty NC, et al. Breast Cancer Index predicts pathological complete response and eligibility for breast conserving surgery in breast cancer patients treated with neoadjuvant chemotherapy. Ann Oncol. 2012; 23(8):2046-2052.
  55. Martin M, Brase JC, Calvo L, et al. Clinical validation of the EndoPredict test in node-positive, chemotherapy-treated ER+/HER2- breast cancer patients: results from the GEICAM 9906 trial. Breast Cancer Res. 2014; 16(2):R38.
  56. Mook S, Schmidt MK, Viale G, et al.; TRANSBIG Constortium. The 70-gene prognosis-signature predicts disease outcome in breast cancer patients with 1-3 positive lymph nodes in an independent validation study. Breast Cancer Res Treat. 2009; 116(2):295-302.
  57. Mook S, Schmidt MK, Weigelt B, et al. The 70-gene prognosis signature predicts early metastasis in breast cancer patients between 55 and 70 years of age. Ann Oncol. 2010; 21(4):717-722.
  58. Nitz U, Gluz O, Christgen M, et al. Reducing chemotherapy use in clinically high-risk, genomically low-risk pN0 and pN1 early breast cancer patients: five-year data from the prospective, randomised phase 3 West German Study Group (WSG) PlanB trial. Breast Cancer Res Treat. 2017 Jun 29. [Epub ahead of print]
  59. Paik S, Shak S, Tang G, et al. A multi-gene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med. 2004; 351(27):2817-2826.
  60. 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):3726-3734. 
  61. Petkov VI, Miller DP, Howlader N, et al. Breast-cancer-specific mortality in patients treated based on the 21-gene assay: a SEER population-based study. NPJ Breast Cancer. 2016; 2:16017. 
  62. Prat A, Parker JS, Fan C, et al. Concordance among gene expression-based predictors for ER-positive breast cancer treated with adjuvant tamoxifen. Ann Oncol. 2012; 23(11):2866-2873.
  63. Rakovitch E, Nofech-Mozes S, Hanna W, et al. A population-based validation study of the DCIS Score predicting recurrence risk in individuals treated by breast-conserving surgery alone. Breast Cancer Res Treat. 2015; 152(2):389-398.
  64. Rakovitch E, Nofech-Mozes S, Hanna W, et al. Multigene expression assay and benefit of radiotherapy after breast conservation in ductal carcinoma in situ. J Natl Cancer Inst. 2017; 109(4). pii: djw256.
  65. Rayhanabad JA, Difronzo LA, Haigh PI, Romero L. Changing paradigms in breast cancer management: introducing molecular genetics into the treatment algorithm. Am Surg. 2008; 74(10):887-890.
  66. Ring BZ, Seitz RS, Beck R, et al. Novel prognostic immunohistochemical biomarker panel for estrogen receptor-positive breast cancer. J Clin Oncol. 2006; 24(19):3039-3047.
  67. Rutgers E, Piccart-Gebhart MJ, Bogaerts J, et al. The EORTC 10041/BIG 03-04 MINDACT trial is feasible: results of the pilot phase. Eur J Cancer. 2011; 47(18):2742-2749.
  68. Saghatchian M, Mook S, Pruneri G, et al. Additional prognostic value of the 70-gene signature (MammaPrint(®)) among breast cancer patients with 4-9 positive lymph nodes. Breast. 2013; 22(5):682-690.
  69. Sestak I, Cuzick J, Dowsett M, et al. Prediction of late distant recurrence after 5 years of endocrine treatment: a combined analysis of patients from the Austrian breast and colorectal cancer study group 8 and arimidex, tamoxifen alone or in combination randomized trials using the PAM50 risk of recurrence score. J Clin Oncol. 2015; 33(8):916-922.
  70. Sestak I, Dowsett M, Zabaglo L, et al. Factors predicting late recurrence for estrogen receptor-positive breast cancer. J Natl Cancer Inst. 2013; 105(19):1504-1511.
  71. Sgroi DC, Sestak I, Cuzick J, et al. Prediction of late distant recurrence in patients with oestrogen-receptor-positive breast cancer: a prospective comparison of the breast-cancer index (BCI) assay, 21-gene recurrence score, and IHC4 in the TransATAC study population. Lancet Oncol. 2013; 14(11):1067-1076.
  72. Solin LJ, Gray R, Baehner FL, et al. A multigene expression assay to predict local recurrence risk for ductal carcinoma in situ of the breast. J Natl Cancer Inst. 2013; 105(10):701-710.
  73. Sparano JA, Gray RJ, Makower DF, et al. Prospective validation of a 21-gene expression assay in breast cancer. N Engl J Med. 2015; 373(21):2005-2014.
  74. Stemmer SM, Klang SH, Ben-Baruch N, et al. The impact of the 21-gene Recurrence Score assay on clinical decision-making in node-positive (up to 3 positive nodes) estrogen receptor-positive breast cancer patients. Breast Cancer Res Treat. 2013; 140(1):83-92.
  75. Stemmer SM, Steiner M, Rizel S, et al. Clinical outcomes in patients with node-negative breast cancer treated based on the recurrence score results: evidence from a large prospectively designed registry. NPJ Breast Cancer. 2017a; 3:33.
  76. Stemmer SM, Steiner M, Rizel S, et al. Clinical outcomes in ER+ HER2 -node-positive breast cancer patients who were treated according to the Recurrence Score results: evidence from a large prospectively designed registry. NPJ Breast Cancer. 2017b; 3:32.
  77. van de Vijver MJ, He YD, van ’t Veer LJ, et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med. 2002; 347(25):1999-2009.
  78. van't Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature. 2002; 415(6871):530-536.
  79. Wittner BS, Sgroi DC, Ryan PD, et al. Analysis of the MammaPrint breast cancer assay in a predominantly postmenopausal cohort. Clin Cancer Res. 2008; 14(10):2988-2993.
  80. Wolff A, Hammond ME, Schwartz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007; 25(1):118-145.
  81. Zhang Y, Schnabel CA, Schroeder BE, et al. Breast cancer index identifies early-stage estrogen receptor-positive breast cancer patients at risk for early- and late-distant recurrence. Clin Cancer Res. 2013; 19(15):4196-4205.

Government Agency, Medical Society, and Other Authoritative Publications:

  1. Blue Cross Blue Shield Association. Gene Expression Profiling for Managing Breast Cancer Treatment. TEC Assessment, 2005; 20(3).
  2. Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group. Recommendations from the EGAPP Working Group: can tumor gene expression profiling improve outcomes in patients with breast cancer? Genet Med. 2009; 11(1):66-73.
  3. Harris L, Fritsche H, Mennel R, et al. American Society of Clinical Oncology 2007 update of recommendations for the use of tumor markers in breast cancer. J Clin Oncol. 2007; 25(33):5287-5312.
  4. Harris LN, Ismaila N, McShane LM, Hayes DF. Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: American Society of Clinical Oncology Clinical Practice Guideline. J Clin Oncol. 2016; 34(10):1134-1150.
  5. Krop I, Ismaila N, Andre F, et al. Use of biomarkers to guide decisions on adjuvant systemic therapy for women with early-stage invasive breast cancer: American Society of Clinical Oncology clinical practice guideline focused update. J Clin Oncol. 2017; 35(24):2838-2847.
  6. NCCN Clinical Practice Guidelines in Oncology™ (NCCN). © 2017 National Comprehensive Cancer Network, Inc. For additional information visit the NCCN website at: http://www.nccn.org/index.asp. Accessed on September 29, 2017.
    • Breast Cancer (V2.2017). Revised April 6, 2017.
  7. Wolff AC, Hammond ME, Hicks DG, et al. Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol. 2013; 31(31):3997-4013.
Websites for Additional Information
  1. American Cancer Society. Breast cancer? Available at: https://www.cancer.org/cancer/breast-cancer.html. Accessed on September 29, 2017.
  2. National Cancer Institute. Breast Cancer Treatment. Available at: http://www.cancer.gov/cancertopics/pdq/treatment/breast/Patient/page1. Accessed on September 29, 2017.
  3. National Library of Medicine. Medical Encyclopedia: Breast Cancer. Available at: http://www.nlm.nih.gov/medlineplus/ency/article/000913.htm. Accessed on September 29, 2017.
Index

Breast Cancer Gene Expression Ratio
Breast Cancer Risk Testing Network
DCIS
EndoPredict
H/I
Insight DX Breast Cancer Profile
MammaPrint
MammaTyper
Oncotype DX
Oncotype DX DCIS
The 76-gene “Rotterdam signature” assay
The 41-gene signature assay
THEROS Breast Cancer Index
Theros H/I

The use of specific product names is illustrative only. It is not intended to be a recommendation of one product over another, and is not intended to represent a complete listing of all products available.

Document History

Status

Date

Action

Revised

11/02/2017

Medical Policy & Technology Assessment Committee (MPTAC) review.

Revised

11/01/2017

Hematology/Oncology Subcommittee review. Revised MN criteria H regarding who may order the test. The document header wording updated from “Current Effective Date” to “Publish Date.” Added new MN statement regarding testing 5 years after initial treatment. Added new test to INV and NMN section. Updated the Rationale and References sections. Updated Coding section with 01/01/2018 CPT changes; deleted 0008M effective 12/31/2017, added 81520 and 81521 effective 01/01/2018.

Revised

08/03/2017

MPTAC review.

Revised

11/03/2016

MPTAC review.

Revised

11/02/2016

Hematology/Oncology Subcommittee review. Updated formatting in Position Statement section. Rearranged the INV and NMN section. Updated the Rationale and Reference sections.

Reviewed

05/05/2016

MPTAC review.

Reviewed

05/04/2016

Hematology/Oncology Subcommittee review. Updated Rationale and Reference sections. Updated Coding section with reinstatement of HCPCS code S3854 effective 07/01/16.

Revised

11/05/2015

MPTAC review. 

Revised

11/04/2015

Hematology/Oncology Subcommittee review. Revised medically necessary criteria regarding tumor size from 4.0 cm to 5.0 cm. Updated Rationale, Reference, and Index sections. Updated Coding section to note 01/01/2016 HCPCS changes, also removed ICD-9 codes.

Revised

05/07/2015

MPTAC review. 

Revised

05/06/2015

Hematology/Oncology Subcommittee review. Added several new tests to the investigational and not medically necessary statement. Updated Rationale and Reference sections. 

 

01/01/2015

Updated Coding section with 01/01/2015 CPT changes.

Revised

05/15/2014

MPTAC review. 

Revised

05/14/2014

Hematology/Oncology Subcommittee review. Updated HER2 testing. Added several new tests to the investigational and not medically necessary statement. Deleted medically necessary criteria “Not a pT4 lesion”. Updated Rationale and Reference sections. Updated Coding section with 07/01/2014 CPT changes.

Revised

05/09/2013

MPTAC review.

Revised

05/08/2013

Hematology/Oncology Subcommittee review. Revised note in position statement regarding equivocal HER test results. Added the PAM50 and BCI tests to the investigational and not medically necessary section. Updated Rationale and Reference sections. 

Revised

11/08/2012

MPTAC review. 

Revised

11/07/2012

Hematology/Oncology Subcommittee review. Added investigational and not medically necessary statements regarding repeat Oncotype DX testing and Oncotype DX testing of multiple tumor sites. Updated Rationale and Reference sections.  Updated Coding section with 01/01/2013 CPT changes.

Revised

05/10/2012

MPTAC review. 

Revised

05/09/2012

Hematology/Oncology Subcommittee review. Added Oncotype Dx DCIS to list of investigational and not medically necessary tests. Added new investigational and not medically necessary statement regarding the use of the gene expression profiling as a technique of managing the treatment of DCIS. Updated Rationale and Reference sections.

Revised

05/19/2011

MPTAC review. 

Revised

05/18/2011

Hematology/Oncology Subcommittee review. Updated Reference section.

Reviewed

07/01/2010

Hematology/Oncology Subcommittee review. Updated reference section.

Revised

05/13/2010

MPTAC review. 

Revised

05/12/2010

Hematology/Oncology Subcommittee review. Added the Insight DX Breast Cancer Profile and THEROS Breast Cancer Index to investigational and not medically necessary section. Updated Reference section.

Reviewed

02/25/2010

MPTAC review. Updated Rationale, Definitions and Reference sections. Updated appendix to request more specific HER2 test data.

Reviewed

02/26/2009

MPTAC review. Added the Mammostrat test to Investigational and Not Medically Necessary section. Updated Rationale and Reference sections.

Revised

03/10/2008

MPTAC review. 

Revised

02/29/2008

Hematology/Oncology Subcommittee review. Revised medically necessary criteria. Updated Rationale, Definitions, and Reference sections. Revised Appendix.

Revised

11/29/2007

MPTAC review. 

Revised

11/28/2007

Hematology/Oncology Subcommittee review. Clarified criteria # 6 to add “(Grade 2)”. Changed tumor size in criteria # 7 from 1-3 cm to 0.5-3.0 cm. The phrase “investigational/not medically necessary” was clarified to read “investigational and not medically necessary.” Updated Rationale and Reference sections.

Revised

09/21/2007

MPTAC review. Updated Rationale and Reference section.

Revised

09/17/2007

Hematology/Oncology Subcommittee review. Changed position statement from investigational/not medically necessary to medically necessary with criteria.  Updated Rationale, Coding, Reference, and Index sections. Added physician attestation sheet.

Reviewed

03/08/2007

MPTAC review. Classification changed from LAB to GENE.

Reviewed

12/07/2006

MPTAC review. Updated Rationale and Reference sections.

Reviewed

12/06/2006

Hematology/Oncology Subcommittee review. No change to position statement.  Updated Rationale and Reference sections.

Reviewed

06/08/2006

MPTAC review. 

Reviewed

06/07/2006

Hematology/Oncology Subcommittee review. Discussion at meeting. References updated.

Reviewed

12/01/2005

MPTAC review. 

Reviewed

11/30/2005

Hematology/Oncology Subcommittee review. Discussion at meeting. No change to position statement.

Revised

04/28/2005

MPTAC review. Revision based on Pre-merger Anthem and Pre-merger WellPoint Harmonization.

Pre-Merger Organizations

Last Review Date

Document Number

Title

Anthem, Inc.

 

 

No previous document

WellPoint Health Networks, Inc.

06/24/2004

2.11.22

Assays of Genetic Expression in Tumor Tissue as a Technique to Determine Prognosis in Patients with Breast Cancer