Patients with HER2-positive metastatic breast cancer who had received prior anti-HER2 therapies continued to experience a progression-free survival benefit with margetuximab and a trend toward overall survival compared with trastuzumab when either agent was combined with chemotherapy, according to updated findings from the phase III SOPHIA trial.
Hope Rugo, MD
Patients with HER2-positive metastatic breast cancer who had received prior anti-HER2 therapies continued to experience a progression-free survival benefit with margetuximab (MGAH22) and a trend toward overall survival (OS) compared with trastuzumab (Herceptin) when either agent was combined with chemotherapy, according to updated findings from the phase III SOPHIA trial.
Data from a prespecified second interim OS analysis presented at the 2019 San Antonio Breast Cancer Symposium (SABCS) showed that at a median follow-up of 15.6 months, the median OS in the intent-to-treat (ITT) population was 21.6 months (95% CI, 18.86-24.05) with margetuximab compared with 19.8 months (95% CI, 17.54-22.28) with trastuzumab plus chemotherapy (HR, 0.89; 95% CI, 0.69-1.13; P = .326).
The second interim OS analysis was based on 270 OS events, up from the 158 OS events at the initial analysis. The final prespecified OS analysis is planned after 385 events have occurred, which is anticipated in late 2020, said lead study author Hope S. Rugo, MD.
“The second interim OS analysis favors margetuximab, although the difference is not yet statistically significant,” said Rugo, who is director, Breast Oncology and Clinical Trials Education, University of California San Francisco Helen Diller Family Comprehensive Cancer Center.
At the latest SOPHIA analysis, which had a data cutoff date of September 10, 2019, the investigator-assessed PFS showed a 29% reduction in risk of disease progression or death. The median PFS was 5.7 months (95% CI, 5.22-6.97) in the margetuximab arm compared with 4.4 months (95% CI, 4.14-5.45) in the trastuzumab arm (HR, 0.71; 95% CI, 0.58-0.86; P = .0006).
“Interestingly there is now an evident plateau in the curve that has appeared over time,” said Rugo, also noting that, “As of December 7 of this year, 20 patients remain on margetuximab compared to 10 patients on trastuzumab, consistent with the plateau now seen on the progression-free survival curve.”
“SOPHIA is the first phase III head-to-head trial to show PFS superiority of a novel antibody compared to trastuzumab,” said Rugo.
MacroGenics, the developer of margetuximab, expects to submit a biologics license application to the FDA before the end of 2019.
Explaining the mechanism of action of the treatment, Rugo said, “Margetuximab is a novel HER2-positive antibody with Fc engineering that alters Fc receptor affinities. “Margetuximab’s Fc engineering was designed to increase IgG1 Fc affinity for the activating Fc gamma receptor CD16A and decrease affinity for the inhibitory Fc gamma receptor CD32B.”
“Margetuximab’s Fc engineering enhances both innate and adaptive immunity, creating a coordinated engagement of HER2-targeted immunity,” added Rugo.
Based on the mechanism of action of margetuximab, prespecified exploratory analyses were conducted to assess the CD16A genotype as a predictor of anti-HER2 efficacy. In the estimated 86% of patients carrying a CD16A 158F allele, the median OS was 23.7 months in the margetuximab arm versus 19.4 months in the trastuzumab arm (HR, 0.79; 95% CI, 0.61-1.04; P = .087).
However, in the 14% of patients who were homozygous for the CD16A-158VV allele, the trastuzumab regimen performed better than the margetuximab regimen. The median OS was 19.7 months in the margetuximab arm compared with 33.3 months in the trastuzumab arm (HR, 1.65; P = .157). In this 69-patient subgroup, “clinical characteristics were unbalanced in favor of the trastuzumab arm, where more patients treated with margetuximab had visceral metastases, brain metastases, more than 2 metastatic lines of therapy, and were older,” explained Rugo.
“Clearly more data are needed; however, at the present time, it seems reasonable to conclude from these data there is no evidence to support a benefit of margetuximab over trastuzumab in the VV homozygous population,” said Rugo.
The updated analysis Rugo presented at SABCS also included investigator-assed response rates. The objective response rate (ORR) was significantly improved with margetuximab at 25.2% versus 13.7% with trastuzumab (P = .0006). The clinical benefit rate was also significantly improved at 48.1% versus 35.6%, respectively (P = .0025).
The ORR in the margetuximab arm comprised a complete response (CR) rate of 1.9%, partial response (PR) rate of 23.3%, stable disease (SD) rate of 53.8%, and progressive disease (PD) rate of 15.0%. The ORR in the trastuzumab arm comprised a CR rate of 1.5%, PR rate of 12.2%, SD rate of 58.5%, and PD rate of 21.1%. The median duration of response was 6.9 versus 7.0 months in the 2 arms, respectively.
Earlier data from the SOPHIA trial were presented during the 2019 ASCO Annual Meeting. At a data cutoff date of October 10, 2018, the median PFS by central blinded analysis (CBA) was 5.8 months in the margetuximab arm compared with 4.9 months in the trastuzumab arm (HR, 0.76; P = .033).2 Investigator-assessed PFS at the cutoff was also longer in the margetuximab arm (median PFS, 5.6 vs 4.2 months; HR, 0.70; P = .001).
At the first interim OS analysis (median follow-up, 9.2 months) the median OS in the ITT population was 18.9 months in the margetuximab arm versus 17.2 months in the trastuzumab arm (HR, 0.95; 95% CI, 0.69-1.31; P = .758).
SOPHIA, included 536 patients (ITT population) with metastatic or unresectable locally-advanced relapsed/refractory HER2-positive breast cancer. All patients had received ≥2 prior anti-HER2 therapies, including pertuzumab (Perjeta) and trastuzumab, and had received 1 to 3 prior lines of treatment in the metastatic setting. Patients could have prior brain metastasis if it was treated and stable.
Baseline characteristics were balanced between the 2 arms, with a median age of 55 years in the margetuximab arm and 56 years in the trastuzumab arm. Ninety-eight percent of patients in each arm had metastatic disease, with 2% in each arm having unresectable locally advanced disease. About 60% of patients in each arm had hormone receptorpositive disease.
The backbone chemotherapy regimens were capecitabine in 27% in each arm, eribulin (Halaven) in about one-fourth, gemcitabine in 12%, and vinorelbine in about one-third of patients.
Fifty-nine percent of patient in the margetuximab arm and 54% in the trastuzumab arm received prior adjuvant and/or neoadjuvant therapy. About one-third in each arm received >2 lines of prior therapy in the metastatic setting. Nearly all patients in each arm had prior taxane-based chemotherapy and about 40% had prior anthracycline-based chemotherapy. Nearly half of each arm had prior endocrine therapy. More than 90% of patients received prior T-DM1 (ado-trastuzumab emtansine; Kadcyla), and about 15% had received lapatinib
Patients were randomized to margetuximab at 15 mg/kg every 3 weeks (n = 266) or trastuzumab (n = 270), using an 8 mg/kg loading dose followed by 6 mg/kg every 3 weeks. All patients also received investigator’s choice of chemotherapy (capecitabine, eribulin, gemcitabine, or vinorelbine) in 3-week cycles.
The sequential primary endpoints were PFS by CBA and OS. Secondary endpoints included investigator assessed PFS and ORR by CBA. A preplanned exploratory secondary endpoint included evaluation of the effect of margetuximab by Fc-gamma receptor alleles (CD16A, CD32A, and CD32B).
The safety analysis was conducted at a data cutoff date of April 2019, and included 264 patients in the margetuximab arm and 266 in the trastuzumab arm. Rugo said the safety data were comparable between the 2 arms and were consistent with previously reported data for the treatments.
Grade ≥3 adverse events (AEs) occurred in 142 (53.8%) patients on the margetuximab arm compared with 140 (52.6%) patients on the trastuzumab arm. Serious AEs occurred in 43 (16.3%) patients on the margetuximab arm compared with 49 (18.4%) patients on the trastuzumab arm.
Infusion-related reactions (IRRs) were more common with margetuximab than with trastuzumab (13% vs 3%) and were mostly grade 1/2 and associated with the first cycle. There were 4 grade 4 IRRs in the margetuximab arm and none in the control arm. All-grade left ventricular dysfunction rates were equivalent across groups, noted Rugo.