Neoadjuvant Letrozole/Entrectinib Shows Limited Activity in Invasive Lobular Breast Cancer

HER2+ Breast Cancer Brain Metastases |

Image Credit: ©Sebastian Kaulitzki – stock.adobe.com

No efficacy-evaluable patients with invasive lobular breast carcinoma (n = 41) achieved a residual cancer burden (RCB) of 0 or 1 following treatment with first-line endocrine therapy combined with entrectinib (Rozlytrek), failing to meet the primary end point of the single-arm phase 2 ROSALINE trial (NCT04551495). 1

Data presented at the 2025 ESMO Breast Congress showed that the rates of RCB II and III after treatment with entrectinib plus letrozole with or without goserelin in the efficacy-evaluable patient population were 61% and 39%, respectively.

In the intention-to-treat (ITT) population (n = 52), which included patients who received fewer than 80% of the planned doses of entrectinib, 1 patient achieved an RBC of 0 or I. The rates of RCB II and III were 50% and 48%, respectively.

“Endocrine therapy plus entrectinib demonstrated limited efficacy in the neoadjuvant setting of lobular breast cancer,” Soraia Lobo-Martins, MD, a medical oncologist and research fellow at Institut Jules Bordet in Brussels, Belgium, stated in the presentation.

Setting the Stage for ROSALINE

Invasive lobular breast cancer comprises approximately 15% of all breast cancer cases and has different clinical and pathological characteristics compared with other subtypes. One of the most frequent genomic events in this disease subtype is a CDH1 alteration, and approximately 90% of invasive lobular breast tumors lack E-cadherin. Notably, pathologic complete response (pCR) rates with endocrine therapy in this subtype are relatively low, even with the addition of CDK4/6 inhibitors. For instance, in the phase 2 neoMONARCH trial (NCT02441946), the pCR rate was 4% among 190 postmenopausal patients with stage I to III hormone receptor–positive, HER2-negative breast cancer who underwent surgery following neoadjuvant treatment with abemaciclib (Verzenio) with or without anastrozole.2

Importantly, in vivo studies with ROS1 inhibitors such as crizotinib (Xalkori) and foretinib (formerly XL880) have shown synthetic lethality between E-cadherin deficiency and ROS1 inhibition.1 The study investigators hypothesized that entrectinib—a potent small-molecule TKI that targets the TRK, ROS1, and ALK tyrosine kinases, which are active in tumors harboring NTRK, ROS1, and ALK gene fusion mutations—would be effective in the first-line invasive lobular breast cancer setting based on previously reported data with this agent in lung cancer.

Outlining the Design of ROSALINE

The investigator-initiated ROSALINE trial enrolled female patients at least 18 years of age with previously untreated stage IIA to IIIA invasive lobular breast cancer who had a tumor size greater than 20 mm, N0 or N1 disease, and an ECOG performance status of 0 or 1. Patients underwent a 28-day screening period, followed by 4 28-day cycles of treatment with entrectinib at 600 mg plus letrozole at 2.5 mg with or without goserelin at 3.6 mg. After neoadjuvant treatment, patients underwent surgery. Notably, entrectinib treatment was required to be interrupted at least 7 days before the planned surgery, letrozole treatment was permitted to be continued until the day of surgery, and goserelin treatment could be initiated prior to study enrollment.

The study’s primary end point was the rate of RCB 0/I per local assessment. Secondary end points included pCR rate per local assessment, overall response rate (ORR) per locally assessed MRI, ORR per mRECIST criteria, and safety. Translational research evaluated genomic aberrations and transcriptional programs, RNA sequencing data, and circulating tumor DNA findings.

Characterizing the Patient Population

Patients were recruited between January 1, 2021, and March 15, 2024. In total, 65 patients were assessed for eligibility, 56 of whom were eligible for enrollment and began treatment. Among enrolled patients, 41 were deemed evaluable. Notably, 15 patients who received fewer than 80% of the planned doses of entrectinib were not deemed evaluable.

The median age of all evaluable patients was 56 years (interquartile range, 48-62). Among these patients, 44% were premenopausal, and 85% had an ECOG performance status of 0. Clinical stage IIA, IIB, and IIIA disease was represented in 37%, 46%, and 17% of patients, respectively. Additionally, 12%, 78%, and 10% of patients had grades 1, 2, and 3 disease, respectively. All patients had estrogen receptor–positive disease, 10% of patients had progesterone receptor–negative disease, and 56% of patients had HER2 1+ or 2+ nonamplified disease per in situ hybridization. A total of 23% of patients had a Ki-67 score above 20%.

Showcasing Additional Efficacy Findings

No patients in the efficacy population achieved a pCR. The rates of stage I, IIA, IIB, IIIA, and IIIC disease were 11%, 30%, 20%, 30%, and 10%, respectively. The MRI-assessed confirmed ORR (cORR) rate was 49%, including best overall responses of complete response (CR; 10%), partial response (PR; 39%), and stable disease (SD; 51%).

In the ITT population, 1 patient achieved a pCR after receiving letrozole and 7 days of entrectinib. The rates of stage I, IIA, IIB, IIIA, and IIIC disease were 10%, 28%, 23%, 28%, and 9%, respectively. The MRI-assessed cORR rate was 48%, including best overall responses of CR (8%), PR (40%), and SD (52%).

Although the threshold for RCB 0/I was not met in stage 1 per the trial’s Simon’s 2-stage design, investigators proceeded with stage 2 of the study due to these ORR data.

Spotlighting Safety Data and Next Steps

Adverse effects (AEs) occurred in all patients in the efficacy population, 22% of which were grade 3 or higher. Serious AEs (SAEs) and AEs of special interest were seen in 7% and 2% of patients, respectively. Dose reductions occurred in 61% of patients, and no patients discontinued treatment.

In the safety-evaluable ITT population (n = 56), AEs occurred in 98% of patients, 25% of which were grade 3 or higher. SAEs and AEs of special interest were reported in 7% and 5% of patients, respectively. Dose reductions and treatment discontinuations occurred in 55% and 18% of patients, respectively.

Notably, no grade 5 AEs were reported. AEs of special interest included cognitive disorders, memory impairment, and congestive heart failure (n = 1 each).

The most common any-grade AEs included constipation (71%), dysgeusia (61%), vertigo (48%), fatigue (43%), arthralgia (38%), nausea (32%), diarrhea (29%), peripheral edema (29%), asthenia (21%), myalgia (21%), dyspnea (20%), dizziness (18%), stomatitis (18%), dysesthesia (16%), hot flush (16%), paresthesia (16%), peripheral sensory neuropathy (16%), rash (13%), balance disorder (11%), decreased appetite (11%), disturbance in attention (11%), muscular weakness (11%), and vomiting (11%). The most common grade 3 or higher AEs were arthralgia (5%), fatigue (4%), rash (4%), vertigo (2%), peripheral edema (2%), and myalgia (2%).

“Despite these results being discouraging, translational analyses to identify mechanisms of resistance and identify biomarkers are ongoing and will be presented at [a later meeting],” Lobo-Martins concluded.

Disclosures: Lobo-Martins reported receiving honoraria and/or advisory board compensation from Roche, Novartis, Pfizer, BMS, AstraZeneca, MSD, and Gilead Sciences; receiving travel grants from Roche, Novartis, Daiichi Sankyo, AstraZeneca, BMS, Pierre Fabre, MSD, Lilly, Pfizer, Sanofi, Amgen, and Gilead Sciences; and participating as a medical research fellow in research studies institutionally funded by AstraZeneca, Novartis, and F. Hoffman-La Roche Ltd to Institut Jules Bordet.

References

1. Lobo-Martins S, Agostinetto E, Duhoux FP, et al. Primary endpoint of ROSALINE: a phase II neoadjuvant study of endocrine therapy (ET) and entrectinib in invasive lobular breast carcinoma (ILBC). Presented at: 2025 ESMO Breast Congress; May 14-17, 2025; Munich, Germany. Abstract 192MO.
2. Hurvitz SA, Martin M, Press MF, et al. Potent cell-cycle inhibition and upregulation of immune response with abemaciclib and anastrozole in neoMONARCH, phase II neoadjuvant study in HR+/HER2- breast cancer. Clin Cancer Res. 2020;26(3):566-580. doi:10.1158/1078-0432.CCR-19-1425