Kathleen N. Moore, MD
Mirvetuximab soravtansine (IMGN853) demonstrated a favorable toxicity profile and encouraging clinical activity in patients with epithelial ovarian cancer, according to a pooled analysis of phase I expansion cohorts that was presented at the 2017 ASCO Annual Meeting.
There were 113 patients with epithelial ovarian cancer in the pooled analysis, including a cohort of 36 patients who met the inclusion criteria for the ongoing, randomized phase III FORWARD I trial (NCT02631876) comparing mirvetuximab soravtansine to investigator’s choice of chemotherapy. The criteria include moderate-to-high folate receptor alpha (FRα) expression and 1 to 3 prior lines of therapy.
The safety profile for the overall pooled population was consistent with previously reported data, with the most common adverse events including diarrhea, fatigue, nausea, and blurred vision; however, these were low grade and well-managed.
In all pooled patients from the phase I analysis, the confirmed overall response rate (ORR) was 30% (95% CI, 22-39), and included 3 complete responses and 31 partial responses. The confirmed ORR was 47% (95% CI, 30-65) in the group eligible for FORWARD I, including 1 complete response and 16 partial responses.
In an interview with OncLive
at ASCO, lead study author Kathleen Moore, MD, assistant professor in the section of gynecologic oncology and director of the Oklahoma TSET Phase I Clinical Trials Program at the University of Oklahoma Health Sciences Center, discussed the potential of mirvetuximab soravtansine in patients with epithelial ovarian cancer.
OncLive: Can you provide an overview of the analyses you presented?
Mirvetuximab soravtansine is the first antibody-drug conjugate in a pivotal phase III trial for patients with ovarian cancer. The abstract that I presented was a pooled analysis of the phase I expansion cohorts. There are 3 expansion cohorts.
We did a pooled analysis, which looked at all of the patients and then we specifically looked at the patients who met eligibility for our open phase III trial. The analysis of the response rate and the PFS that were reported in the first expansion cohort informed the design of the phase III study. We wanted to show consistency of those findings with a larger data set.
This analysis was done to show that the high response rate in the patient population we identified was real and justifies the design of the ongoing [phase III] trial.
Can you give some background into the mechanism of action of mirvetuximab and then provide an overview of the phase III trial design?
I describe antibody-drug conjugates to patients as arrows or bombs. The head of the arrow is targeted to something that is ideally tumor specific, such as a receptor or protein that is only present on the tumor in order to know the expression on normal tissue.
The stem of the arrow is what is called a linker. Linkers are elegantly designed structures that hold the head of the arrow to the tail of the arrow. The tail of the arrow is conjugated with highly potent molecules of chemotherapy. In the case of mirvetuximab, this is a drug called DM4, which is a microtubule toxin. It's an inactive drug, so the linker is important because it needs to keep the entire molecule together until it hits its target.
In the case of mirvetuximab, the target is FRα, which is widely overexpressed on high-grade epithelial ovarian cancer. However, 80% of cases overexpress this and about 60% to 65% have moderate-to-high expression, which is what we’re looking for in the phase III trial. Once it binds, it’s like a Trojan horse. The cell thinks it’s a friend through endocytosis, then it’s led into the cell and the linker releases the molecules of chemotherapy. If the cell is sensitive to DM4, the cell will die. You will get apoptosis and the molecules can diffuse into surrounding cells. You get 2 chances to kill the cell you’ve landed on and some surrounding cells.
In the phase I trial, during the dose expansion in platinum-resistant patients, we had response rates of over 40% in the group of patients who had 1 to 3 prior lines of chemotherapy and moderate-to-high expression of FRα. If you look at single-agent cytotoxic therapy as a comparison, the response rate is in the 15% range, making it a significantly higher response rate than we normally see.
The median PFS was almost 7 months, which is also significant compared with single-agent cytotoxic therapy that normally has a median PFS of around 4 months. That was the signal that was seen in the first expansion cohort and was very consistent. We have a 47% response rate and a 6.7-month median PFS in the pooled cohort.