SOFT Trial Identifies Index Risk Scores as Prognostic Tools for Premenopausal Patients with HR+ BC Receiving Adjuvant ET

Ruth O’Regan, MD

In an analysis of the SOFT trial (NCT00066690), the Breast Cancer Index (BCI) test accurately identified premenopausal women undergoing adjuvant endocrine therapy for early-stage, hormone receptor (HR)–positive breast cancer who would derive benefit from adding ovarian function suppression (OFS).¹

Ruth O’Regan, MD, chair of medicine and Charles A. Dewey Professor at the University of Rochester, presented the findings during the 2022 San Antonio Breast Cancer Symposium (SABCS). “This is the first genomic assay to demonstrate benefit from ovarian suppression, supporting additional clinical utility and value and validation of BCI and premenopausal women,” she said.

SOFT, a landmark international, three-arm, non-blinded, randomized clinical trial, included 3066 premenopausal women with HR-positive breast cancer. Patients were randomized 1:1:1 to receive tamoxifen alone, tamoxifen plus OFS, or exemestane plus OFS, all for 5 years. Data from that trial showed a 3% improvement in 12-year freedom from distant recurrence (DRFI) with exemestane plus OFS compared with tamoxifen alone (HR, 0.75; 95% CI, 0.59-0.97).

However, OFS can be toxic, and patients can struggle with compliance. Therefore, it is crucial to identify which patients are most likely to benefit, O’Regan said. Data from SOFT showed that women younger than 35 years appear to derive benefit from OFS following chemotherapy. Furthermore, investigators have developed a continuous composite that incorporates risk value age, nodal status, tumor size, grade, estrogen receptor (ER) status, progesterone receptor status, and Ki-67 that can assist with estimating the individual risk-based benefit of escalating endocrine therapy for HER2-negative patients.

“However, today, no genomic biomarker has been available to predict the benefit of errands of ovarian suppression premenopausal patients,” O’Regan said. “This is critical because of not just the short-term toxicities, but also long-term toxicities, which may not already be fully established.”

To develop such a biomarker, O’Regan and her colleagues evaluated the predictive and prognostic ability of the BCI test in patient tumor samples (n = 1717) collected from the SOFT trial.

The BCI is a genomic assay that incorporates molecular grade index (MGI) and the HOXB13/IL17BR (H/I) index ratio. In ER-positive breast cancer, the BCIN+ further incorporates tumor size and grade for N1 disease. BCI is prognostic for patents with N0 disease who received endocrine therapy with or without chemotherapy and BCIN+ is predictive for those with N1 disease.

“BCI has been found to be prognostic in determining individualized risk of late recurrences and also cumulative overall distant recurrence from year 0 to year 10,” O’Regan said.

The BCI predictive part is focused on the (H/I) ratio and is exclusively on the ability of the (H/I) AI to assess question signaling pathways. “What it does is report the likelihood of benefit from extended endocrine therapy as a ‘Yes,’ if the BCI (H/I) is high or ‘No’ if the BCI (H/I) is low,” O’Regan said.

The primary study objective was determining whether BCI (H/I) status was predictive for benefit with exemestane plus OFS vs tamoxifen. Investigators hypothesized that BCI (H/I)-high status would be predictive for OFS benefit, while BCI (H/I) low status would not.

In the data presented at SABCS, investigators assessed tumor samples collected from 573 patients assigned to tamoxifen, 551 assigned to tamoxifen plus OFS, 563 assigned to exemestane. The median follow-up was 12 years.

In the BCI analysis cohort (n = 1687), half (50.1%) of patients had grade 2 tumors, 25.4% had grade 1, 22.9% had grade 3, and tumor grade was unknown in 1.5%. Patient age ranged from younger than 35 years (11.3%) to older than 50 years (10.6%). More than 65% of patients had pN0 nodal status, 25.3% had pN+ 1 to 3 status, and 9.0% had pN+ 4+ status. Tumor size was equal to or less than 2 cm in 64.1% of patients, greater than 2 cm in 34.4%, and was unknown in 1.5%.

For HER2 status, 11.7% of patients were positive. Most patients (53.3%) received chemotherapy. O’Regan said the BCI analysis cohort was representative of the parent SOFT trial.

Investigators determined that 58% of cancers were BCI (H/I) low and 42% were BCI (H/I) high. There was a significant treatment by biomarker interaction for exemestane plus OFS vs tamoxifen (P <.01). The interaction was not as strong for tamoxifen plus OFS vs tamoxifen (P = .16).

In the overall HR-positive cohort, those with BCI (H/I)-high tumors, patients did not derive an absolute benefit for exemestane plus OFS vs tamoxifen (HR, 1.03; 95%, 0.70-1.53) or tamoxifen plus OFS vs tamoxifen (HR, 1.05; 95% CI, 0.72-1.54). The absolute benefit was –0.4% for exemestane plus OFS and –1.2% for tamoxifen plus OFS.

In contrast, investigators observed an absolute benefit of 11.6% for exemestane plus OFS vs tamoxifen (HR, 0.48; 95%, 0.33-0.71) in the BCI (H/I)-low group. For tamoxifen plus OFS vs tamoxifen, the absolute benefit was 7.3% (HR, 0.69; 95% CI 0.48-0.97).

In patients with HR-positive, HER2-negative tumors, BCI (H/I) was again predictive in the low group. The absolute 12-year benefit was 13.2% for exemestane plus OFS vs tamoxifen (HR, 0.39; 95% CI, 0.25-0.60) and 7.4% for tamoxifen plus OFS vs tamoxifen (HR, 0.64; 95% CI, 0.44-0.93).

In the BCI (H/I) group, the absolute benefit was –0.2% for exemestane plus OFS vs tamoxifen (HR, 1.03; 95% CI, 0.64-1.68) and –6.8% for tamoxifen plus OFS vs tamoxifen (HR, 1.37; 95% CI, 0.89-2.13).

Investigators also found that predictive performance was generally consistent across subgroups, including nodal status, prior chemotherapy, and age. Furthermore, consistent with data reported from the SOFT trial, the benefit derived from tamoxifen plus OFS was smaller than the benefit derived from exemestane plus OFS.

“BCI risk scores were prognostic in premenopausal women with HR-positive tumors receiving adjuvant endocrine therapy [with or without] chemotherapy in that higher BCI risk scores were associated with the worse outcome,” O’Regan said. “The BCI (H/I) was predictive of ovarian suppression benefit. However, contrary to our study hypothesis, the BCI (H/I)-low group consistently derived clinically meaningful benefit while the BCI (H/I)-high group did not. We’re not sure what these results mean, but it does point to potential differences in tumor biology underlying the ovarian suppression response.”

_REFERENCE

_{O’Regan R, Zhang Y, Fleming GF, et al. Evaluation of the Breast Cancer Index in premenopausal women with early-stage HR+ breast cancer in the SOFT trial. Presented at: 2022 San Antonio Breast Cancer Symposium; December 6-10, 2022; San Antonio, TX. Abstract GS1-06.}