Skip to main content

Roche's Ipatasertib Shows No PFS Benefit in Phase III, Biomarker-Guided TNBC Trial

Premium

NEW YORK – Roche's investigational AKT inhibitor ipatasertib in combination with paclitaxel failed to significantly hold back disease progression compared to just paclitaxel in a Phase III trial involving triple-negative breast cancer patients with certain mutations, according to data presented at the San Antonio Breast Cancer Symposium last week.

Researchers led by Rebecca Dent, associate professor at the National Cancer Center in Singapore, presented the results at the meeting from cohort A of the IPATunity130 trial. Patients in this cohort had advanced or metastatic TNBC with PIK3CA, AKT1, or PTEN alterations. Approximately 35 percent of TNBC patients harbor such aberrations in their tumors.

The late-stage failure comes after ipatasertib plus paclitaxel improved progression-free survival in advanced TNBC patients compared to just paclitaxel in the Phase II LOTUS trial. Dent was also involved in that study and noted during her SABCS presentation that progression-free survival benefits were particularly robust in the subgroup of patients with PIK3CA, AKT1, or PTEN-altered tumors, and median overall survival also appeared to favor this molecularly defined subset on ipatasertib-paclitaxel.  

Specifically, in the LOTUS trial, 42 TNBC patients with PIK3CA/AKT1/PTEN-alterations and on ipatasertib-paclitaxel saw a median overall survival of 25.8 months compared to 22.1 months for those on paclitaxel. When researchers considered the entire study cohort of 124 advanced TNBC patients, median overall survival was also 25.8 months compared to 16.9 months in the comparator arm. Progression-free survival was 9 months versus 4.9 months in the biomarker-defined group receiving ipatasertib-paclitaxel versus paclitaxel, and 6.2 months versus 4.9 months in the overall population, respectively.

Based on the encouraging findings from this earlier Phase II trial, Dent and her colleagues decided to further study this subpopulation in a larger Phase III trial. In the 255-patient IPATunity130 trial, however, researchers saw a less pronounced effect on median progression-free survival. As of the data cutoff on May 7, and at a median follow-up of 8.3 months, patients with PIK3CA/AKT1/PTEN alterations had a median progression-free survival of 7.4 months on the ipatasertib-containing regimen compared to 6.1 months for the placebo arm, which was not a statistically significant improvement.

In terms of the overall response rate, in the ipatasertib-paclitaxel arm, 39 percent of patients saw their tumors shrink, compared to 35 percent in the paclitaxel-placebo arm.

Researchers used Foundation Medicine's FoundationOne CDx next-generation sequencing panel in this study to screen patients' tissue samples for alterations in the three genes. In order to be considered biomarker-positive and eligible for the study, TNBC patients could have one or more of these biomarkers: activating mutations in PIK3CA and AKT1, and loss-of-function alterations in PTEN.

"This is one of the first trials where [patients] were able to get the NGS assay, especially in a first-line setting," Dent noted. Even though for this analysis patients with alterations in any of the three genes were grouped as "biomarker positive," in the future Dent's team will explore whether patients' outcomes differ if they have alterations in PIK3CA, AKT1, or PTEN. The researchers also conducted circulating tumor DNA analysis and will compare the concordance between liquid biopsy and tissue-based analysis of these biomarkers, she added.

When researchers considered outcomes in specific subgroups within this study, Dent noted there was an "interesting interaction" between patients' PD-L1-positive status and slightly longer median progression-free survival on the ipatasertib-paclitaxel arm. However, she said this observation needed to be followed up in a larger study.

While IPATunity130 did not show significant progression-free survival improvement with ipatasertib-paclitaxel for this patient subgroup, Dent said the "story is only beginning" for exploring the activity of AKT inhibitors in TNBC. Trials studying other AKT inhibitors, like AstraZeneca's capivasertib, in a broader TNBC population, not just a biomarker-specific group, may offer further insights into the activity of this class of drugs, Dent said.

Her team is still analyzing overall survival data in the IPATunity130 trial, and Dent noted that despite the lackluster progression-free survival data, ipatasertib-paclitaxel may still show an overall survival advantage for patients with PIK3CA, AKT1, or PTEN alterations.

This trial also "emphasizes the heterogeneity" of TNBC, Dent said, because even the control arm data was different from previous studies using the same chemotherapy. Patients in the IPATunity130 paclitaxel-placebo arm saw a median progression-free survival that was almost double what was observed in other TNBC studies, like KEYNOTE-335, which also used paclitaxel in the control arm, she said.

"It's important to recognize that TNBC is very heterogeneous and it's a diagnosis of exclusion," Dent explained. Although TNBC is defined as tumors that aren't expressing estrogen, progesterone, or HER2 receptors, these tumors still rely on signals from various molecular pathways to survive.

"[TNBC] is an overarching term to describe basically a breast cancer that has several different pathways that are upregulated," she said. "AKT is also a complex pathway and there are different feedback loops and different resistance mechanisms that have to be considered."

She also said looking deeper into the drug's interactions with other potentially predictive and prognostic biomarkers, like PD-L1 expression, could help researchers better understand how the heterogeneity of TNBC impacts patients' ability to respond to AKT inhibitors.

"There's obviously an interaction [with PD-L1] and it's important [to study] because we don’t even really understand what AKT inhibition does," Dent said, adding that most likely AKT inhibition remodels the tumor microenvironment. If so, "this interaction with PD-L1 is really important."