LSZ102 was found to be well tolerated and demonstrate clinical activity in combination with either ribociclib or alpelisib in patients with estrogen receptor–positive breast cancer who have progressed on endocrine therapy.
Komal Jhaveri, MD
The oral selective estrogen receptor degrader (SERD) LSZ102 was found to be well tolerated and demonstrate clinical activity in combination with either ribociclib (Kisqali) or alpelisib (Piqray) in patients with estrogen receptor (ER)—positive breast cancer who have progressed on endocrine therapy, according to findings of a phase 1/1b trial that were presented during the 2020 ESMO Breast Cancer Virtual Meeting.1
In the 3-arm study, investigators evaluated LSZ102 alone (arm A), in combination with ribociclib (arm B), and in combination with alpelisib (arm C). In arm A, single-agent LSZ102 led to an objective response rate (ORR) of 1.3%, a clinical benefit rate (CBR) of 9.1%, and a median progression-free survival (PFS) of 1.8 months (95% CI, 1.7-2).
In arm B, the combination of LSZ102 plus ribociclib elicited a 15.8% ORR and a CBR of 35.5%; the median PFS was 6.2 months (95% CI, 4.4-6.4). When LSZ102 was combined with alpelisib, the ORR was 5.4%, the CBR was 18.9%, and the median PFS was 3.5 months (95% CI, 1.8-5.5).
"This is the first clinical report of an oral SERD in combination with either a CDK4/6 inhibitor or an alpha-specific PI3K inhibitor. LSZ102 is a safe and relatively well-tolerated agent both alone and in combination with ribociclib or alpelisib, with gastrointestinal-related toxicities being the most common toxicity across all 3 arms," said lead study author Komal Jhaveri, MD, FACP, a medical oncologist at Memorial Sloan Kettering Cancer Center, in a presentation during the meeting.
"Downregulation of the ER protein was observed across all arms. Additionally, exploratory analyses reveal the clinical activity for both ribociclib and alpelisib combinations with LSZ103 that was observed in heavily pretreated patients with ER-positive breast cancer, regardless of mutational status in [circulating tumor] DNA at baseline."
The combination of the ER pathway can be targeted by endocrine therapies, Jhaveri explained. Moreover, endocrine-resistance pathways identified in ER-positive breast cancer include the cyclin D-CDK4/6-retinoblastoma protein pathway or PI3K/AKT/mTOR pathway. Additional resistance mechanisms involve ESR1 mutations, which drive ER-dependent transcription and proliferation in the absence of estrogen.
Currently, fulvestrant (Faslodex) is the sole FDA-approved SERD, which has demonstrated a PFS and overall survival benefit when used in combination with ribociclib, and a PFS benefit when added to alpelisib.
Fulvestrant is administered via intramuscular injection; however, LSZ102 is an oral SERD, which could potentially achieve higher systemic exposures and lead to enhanced efficacy and potential clinical activity against ESR1 mutations, Jhaveri added. Preclinical models have shown that LSZ102 has synergistic activity with ribociclib and also with alpelisib in ER-positive breast cancer.2,3 Earlier findings of this trial also showcased activity in arm A and B, Jhaveri noted.4
In the 3-arm, open-label, phase 1/1b CLSZ102X2101 (NCT02734615) trial, patients with pre- or postmenopausal, histologically confirmed, ER-positive breast cancer were enrolled. In all arms, patients were 18 years or older with an ECOG performance status of 0 or 1. Patients were required to have demonstrated disease progression on endocrine therapy for locally advanced disease that was not amendable to curative therapy or metastatic disease; or recurrence either during adjuvant therapy or within 1 year of completion. Prior treatment with fulvestrant or CDK4/6 inhibitors was permitted.
Specifically, to be treated with the combination of LSZ102 and alpelisib, patients could not have received prior therapy with PI3K, mTOR, or AKT inhibitors. Patients with and without PIK3CA mutations were permitted.
As mentioned, the study comprised 3 arms. In arm A (n = 78) investigators administered LSZ102 at 200 mg to 900 mg; in arm B (n = 76), 200 mg to 600 mg of LSZ102 plus 200 mg to 600 mg of ribociclib; and in arm C (n = 39), 300 mg to 450 mg of LSZ102 and 200 to 300 mg of alpelisib. LSZ102 was tested daily alone or in combination with ribociclib daily on a 3-week-one/1-week-off schedule or continuously, while alpelisib was given daily.
The primary end points were safety and tolerability, as well as identification of the recommended dose for expansion. Secondary end points were preliminary antitumor activity, pharmacokinetics, and pharmacodynamics (PD) with ER via immunohistochemistry (IHC). Exploratory outcome measures were PD via serial circulating tumor DNA (ctDNA) and response/resistance by baseline ESR1 or PIK3CA status via ctDNA or biopsy.
Baseline characteristics were similar between the arms. The median age was 59.0 years in arms A (range, 30-70) and B (range, 33-79), and 55 (range, 36-74) in arm C. The ECOG performance status was 0 in 53, 57, and 29 patients in arms A, B, and C, respectively. Twenty-five, 19, and 10 patients had an ECOG performance status of 1 in arms A, B, and C, respectively. The median number of prior lines of antineoplastic therapy in the locally advanced/metastatic setting was 4 in arms A and B, and 3 in arm C. The median number of prior lines of chemotherapy in all arms was 1.
Across all 3 arms, 57.2% of patients received prior fulvestrant, 55.3% of patients received prior treatment with a CDK4/6 inhibitor, and 67.96% of patients received chemotherapy.
Zero patients in arm A had treatment ongoing at the time of data cutoff, which was July 2019, 13 patients (17.1%) in arm B, and 8 patients (20.5%) in arm C had ongoing treatment. All patients in arm A discontinued treatment, compared with 63 patients (82.9%) in arm B and 31 patients (79.5%) in arm C. In arm A, most patients discontinued treatment (n = 71; 91%) due to disease progression, compared with 58 patients (76.3%) in arm B and 25 patients in arm C (64.1%). There were 3 deaths overall, all of which occurred in arm C, and were unrelated to study treatment.
Regarding safety, results showed that the most frequent adverse events (AEs) across all 3 arms were gastrointestinal related, consisting of nausea, vomiting, and diarrhea. In arm B, grade 3 AEs included neutropenia (13.2%) and increased aspartate aminotransferase (AST; 3.9%) and were likely driven by ribociclib. In arm C, grade 3 hyperglycemia (10%), skin rashes (15.4%) and were likely driven by alpelisib.
The number of dose-limiting toxicities in arms A, B, and C, were 5, 3, and 7, respectively. In arm A, this was comprised of diarrhea (n = 2), vomiting (n = 1), increased alanine aminotransferase (n = 1), increased AST (n = 1). In arm B, dose-limiting toxicities were febrile neutropenia, sepsis, and decreased appetite (n = 1 each). In arm C, this comprised diarrhea (n = 1), stomatitis (n = 1), hyperglycemia (n = 2), and maculo-papular rash (n = 3).
To determine the recommended dose for expansion for arm C, 4 dose levels were explored: LSZ102 at 300 mg plus alpelisib at 200 mg (n = 12), LSZ102 at 450 mg plus alpelisib at 200 mg (n = 13), LSZ102 at 300 mg plus alpelisib at 300 mg (n = 12), and LSZ102 at 300 mg plus alpelisib at 250 mg (n = 2). Five DLTs were identified at the 300-mg LSZ102 dose plus 300 mg of alpelisib and was determined to not be tolerable.
Based on these data, the recommended dose was determined to be LSZ102 at 300 mg daily plus 250 mg of alpelisib daily. After the data cutoff, an additional 4 patients received this dosage of the combination.
Additional findings showed that the ER protein is downregulated by LSZ102 across all arms. Paired biopsies were collected at screening as well as day 15 of cycle 1, and were analyzed for ER through IHC. These analyses showed a trend of dose dependent ER protein loss in patients who were treated with LSZ102 alone. Moreover, adding either ribociclib or alpelisib to LSZ102 did not impact downregulation of ER protein.
In the ctDNA analysis, investigators noted that the genomic landscape of heavily pretreated patients is dominated by ESR1, PIK3CA, and TP53 mutations. These abnormalities were reported in all 3 arms. However, these mutations did not correlate with response, and are not enriched upon progression in patients with matched baseline and end-of-treatment samples, Jhaveri noted.
Through exploratory analyses, it was also observed that pairing LSZ102 with either ribociclib or alpelisib led to similar outcomes, regardless of the mutational status in baseline ctDNA. In arm B, patients with baseline ESR1 wild-type status had an ORR of 9.38%, a CBR of 31.25%, and a median PFS of 6.2 months. For ESR1-mutant patients, the ORR was 13.79%, the CBR was 31.03%, and the median PFS was 5.6 months (HR, 0.99; 0.54-1.79). For patients with ESR1 undetectable ctDNA, the ORR was 33.33%, the CBR was 53.33%, and the median PFS was 6.5 months (HR, 0.43; 95% CI, 0.19-1.02).
The exploratory analysis also stratified patients by baseline PIK3CA status. In PIK3CA wild-type patients, the ORR, CBR, and median PFS were 7.14%, 14.29%, and 3.4 months, respectively. For patients who had PIK3CA-mutant tumors, the ORR was 6.67%, the CBR was 20%, and the median PFS was 3.5 months (HR, 0.64; 95% CI, 0.26-1.6). In patients with PIK3CA undetectable ctDNA, the ORR was 0%, the CBR was 25%, and the median PFS was 15.3 months (HR, 0.15; 95% CI, 0.02-1.16).
Incoming data suggest relative enhanced activity of the combination in PIK3CA-mutant patients, Jhaveri noted.
Saverio Cinieri, MD, clinical director, director of the Breast Unit at Ospedale Perrino in Brindisi BR, Italy, who also served as a discussant on the study during the virtual meeting, stated that during the COVID-19 era, the use of oral therapies, such as SERDs, will be even more necessary to limit access to hospitals for patients.
"The overcoming of a molecule like the intramuscularly administered fulvestrant goes in this direction," he concluded. "The clinical efficacy and the biomolecular profile of LSZ102 seems to be able to meet these real needs."