Treatment Tailoring Possible With Gene Expression Profiling

Joseph A. Sparano, MD

Each gene expression-based test available for risk prediction in patients with breast cancer has unique properties and they are not interchangeable, placing importance on the clinical studies used to validate the clinical utility of each assay, according to a presentation by Joseph A. Sparano, MD, at the 36th Annual Miami Breast Cancer Conference.

The 5 tests incorporated into the National Comprehensive Cancer Network (NCCN) guideline for invasive breast cancer are Oncotype DX, MammaPrint, Breast Cancer Index (BCI), Prosigna (PAM 50), and Endopredict.¹ Each of these assays examines different molecular characteristics and varying numbers of genes. The Oncotype DX test uses the expression of 21 genes, MammaPrint examines 70 genes, Prosigna utilizes a 50-gene profile, BCI examines 7 key genes and ratios, and EndoPredict is a 12-gene panel.

As laboratory-developed tests, these assays did not require FDA approval before being marketed; however, a few of the assays have obtained 510(k) clearance, specifically MammaPrint and Prosigna. NCCN treatment guideline recommendations for each of these tests have been based on the level of clinical evidence demonstrated in prospective validation, registry, population-based, and clinical trials.

Prospective Clinical Trials of Gene Expression Tests

“There are critical issues in biomarker development and validation,” said Sparano, professor of Medicine & Women’s Health, Albert Einstein College of Medicine, Montefiore Medical Center. “First and foremost is analytical viability, secondly is clinical validity—and many of the biomarkers available check off these boxes for higher analytics and clinical validity—but the real test for a biomarker is clinical utility, that is, does the result of the biomarker change treatment and do patients benefit from that change.”Each test has undergone extensive validation with NCCN level 1 to 2A evidence demonstrating their prognostic value for patients with breast cancer. Of the available tests, the guideline only lists Oncotype DX has showing predictive value, based on findings from the large TAILORx trial.

TAILORx included more than 10,000 patients with hormone receptor (HR)-positive, HER2-negative, axillary node-negative breast cancer.² Patients were stratified by risk using the Oncotype DX assay, with 9719 having follow-up information available for analysis. Overall, 69% (n = 6711) had an intermediate recurrence score of 11 to 25; 17% (n = 1619) had a score of 10 or lower; and 14% (n = 1389) had a score of 26 or higher.

Those in the low-risk group received endocrine therapy alone while patients in the high-risk group were treated with the combination of chemotherapy and endocrine therapy. Patients in the intermediate group were randomized to receive either endocrine therapy alone (n = 3399) or chemotherapy plus endocrine therapy (n = 3312). Endocrine therapy most commonly consisted of an aromatase inhibitor for postmenopausal women and tamoxifen alone or with an aromatase inhibitor for premenopausal women.

After a median of 7.5 years of follow-up, endocrine therapy alone was noninferior to chemotherapy plus endocrine therapy for invasive disease-free survival in the intermediate group (hazard ratio, 1.08; 95% CI, 0.94-1.24; P = .26). Moreover, freedom from recurrence at a distant site was similar between the 2 groups (hazard ratio, 1.10; 95% CI, 0.85-1.41; P = .48).

“We wanted to make sure there wasn’t a subgroup of patients who were benefiting from chemotherapy, so we looked carefully and found no interaction between clinical risk, tumor size, and grade, but we did find interaction between age and chemo benefit, specifically age, recurrence score, and chemo benefit,” said Sparano.

In those 50 years of age or younger, there was a benefit for adding chemotherapy in the intermediate risk group. In the group with a recurrence score of 16 to 20, there was a 1.6% difference between the arms at 9 years. In the 21 to 25 recurrence score group, 6.5% more patients recurred at 9 years in the endocrine alone arm compared with the chemotherapy plus endocrine arm (86.9% vs 93.4% freedom from recurrence rates).

Based on these findings, the NCCN guideline listed Oncotype DX as a preferred assay, giving it a level 1A rating for determining an ability to forego chemotherapy in patients with a recurrence score of 0 to 25, with a notation that patients younger than 50 may still benefit from chemotherapy. Additionally, the guideline lists the test as level 1B for determining the need for chemotherapy in patients with a score >25.

The MINDACT trial³ showed similar findings for the tailoring of adjuvant therapy for women with early-stage breast cancer, Sparano noted. For this study, those determined to have clinically high risk but low genetic risk using the 70-gene MammaPrint test were randomized to receive chemotherapy or no chemotherapy, with a 1.5 percentage point difference between the 2 groups for survival without distant metastasis (hazard ratio, 0.78; 95% CI, 0.50-1.21; P = .27).

Although not incorporated into the guideline as a predictive test, Sparano noted that the BCI test has shown utility at predicting the need for extended adjuvant therapy in the MA.17 study.⁴ The test, which looks at the gene expression ratios of HOXB13 to IL17BR (H/I), showed that patients with a high H/I were more likely to benefit from extended adjuvant therapy compared with a low H/I score.

Discordance Between Assays

In the high H/I group, there was a 16.5% reduction in the risk of recurrence with extended adjuvant therapy compared with 5 years of therapy (P = .007). In the low H/I group, those who received 10 years of therapy had similar outcomes as those receiving 5 years of therapy. Moreover, there was an association between high H/I and added benefit with letrozole (Femara; P = .03).The gene expression assays provide valuable prognostic information for patients with HR-positive, HER2-negative breast cancer, especially those with node-negative and low-volume node-positive disease; however, findings from one test cannot be applied to another, Sparano cautioned. When each assay was compared with Oncotype DX across studies there were varying levels of discordance between each of the assays (Table). This discordance ranged from 40% to 60%, said Sparano.

These assays are not interchangeable, there's a lack of concordance in risk classification,” he added. “If you're going to use one of these assays, you need to use one that you feel comfortable with and that is supported by a level of evidence that you feel comfortable with.”

In the OPTIMA study specifically,5 which included N0 patients with tumors ≥30 mm (19%) and those with N1 disease, the percentage of patients labeled “low risk” differed significantly between tests. Oncotype DX labeled 82.1% of patients as low risk, while Prosgina placed 36.1% and MammaPrint placed 61.4% of patients in this category.

“When you look across multiple studies comparing the 21-gene recurrence score with other assays, there’s only about a 50% concordance, so it’s almost like a flip of the coin for risk stratification with this assay versus other assays,” said Sparano. “If the point of using an assay is to spare the use of chemotherapy, then the use of the 21-gene assay will classify fewer patients as requiring chemotherapy and it is also associated with the highest level of evidence that that group will benefit from chemotherapy.”

References

1. Breast Cancer, version 4.2018. NCCN website. www.nccn.org/professionals/physician_gls/pdf/breast.pdf. Accessed March 7, 2019.
2. Sparano JA, Gray RJ, Makower DF, et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med. 2018; 379:111-121.
3. Cardoso F, van't Veer LJ, Bogaerts J, et al. 70-Gene Signature as an Aid to Treatment Decisions in Early-Stage Breast Cancer. N Engl J Med. 2016; 375:717-729.
4. Sgroi DC, Carney E, Zarrella E, et al. Prediction of late disease recurrence and extended adjuvant letrozole benefit by the HOXB13/IL17BR biomarker. J Natl Cancer Inst. 2013.105(14):1036-1042.
5. 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): djw050

<<< 2019 Miami Breast Cancer Conference