NEW YORK – In an exploratory biomarker analysis of the Phase III ASCENT study presented during the San Antonio Breast Cancer Symposium on Thursday, investigators explored whether differing levels of Trop-2 expression on the surface of breast cancer cells or the presence of germline BRCA1/2 gene mutations might influence the efficacy of sacituzumab govitecan-hziy (Gilead/ Immunomedic's Trodelvy).
Despite the fact that the antibody drug conjugate is designed to deliver an anti-tumor payload to triple-negative breast tumors specifically by binding to the cell-surface marker Trop-2, the drug's US Food and Drug Administration-approved indication doesn't require that patients' tumors express a certain level of Trop-2 to be treatment eligible. The agency approved the drug in April for the treatment of metastatic TNBC regardless of biomarker status.
The FDA granted accelerated approval to sacituzumab govitecan-hziy on the basis of TNBC patients' objective response rates in a Phase II basket study, after which the confirmatory Phase III clinical trial, dubbed ASCENT, was initiated to further confirm the drug's benefit. Results from ASCENT, presented in September, demonstrated that patients treated with sacituzumab govitecan-hziy experienced improved progression-free survival and overall survival compared with patients who received their physician's choice single-agent chemotherapy. Specifically, patients on sacituzumab govitecan-hziy experienced a median progression-free survival of 5.6 months and a median overall survival of 12.1 months versus 1.7 months and 6.7 months, respectively, for those on single-agent chemo.
Although the 529 patients enrolled on the ASCENT trial were stratified based on the number of lines of chemotherapy they had received, their geographic regions, and whether their cancers had metastasized to the brain, there was no such stratification written into the trial design as far as biomarkers were concerned. As such, the biomarker analysis that Sara Hurvitz of the University of California, Los Angeles presented on Thursday was exploratory in nature, based on patients' archival biopsies or surgical specimens collected within ASCENT.
Hurvitz explained during a presentation at SABCS the rationale behind the exploratory biomarker assessment. Previously conducted translational research has shown that roughly 88 percent of patients with TNBC have moderate-to-strong Trop-2 expression based on immunohistochemistry testing, and most of these patients have tumors that express Trop-2 in over 50 percent of their tumor cells. Previous research in breast cancer xenografts has also indicated that tumors with high Trop-2 expression may derive greater benefit from sacituzumab govitecan-hziy.
To assess levels of Trop-2 expression in patients' tumors, ASCENT investigators used IHC to assign an H-score between 0 and 300 based on staining percentages. For the purposes of this analysis, "Trop-2-low" status was given to tumors with H-scores below 100; "Trop-2-medium" status was assigned to tumors with H-scores between 100 and 200; and "Trop-2-high" status was assigned to tumors with H-scores between 200 and 300.
"This is a pretty ubiquitously expressed protein," Hurvitz said, noting that it's likely there weren't many patients enrolled in the trial whose tumors did not express any Trop-2 at all.
Among patients without brain metastases evaluable for the biomarker, over half in the sacituzumab govitecan-hziy and comparator treatment arms had Trop-2-high expression levels. Across all three Trop-2 expression level subgroups, progression-free survival was longer for patients receiving sacituzumab govitecan-hziy versus chemo.
Among Trop-2-high patients, the median progression-free survival was 6.9 months with sacituzumab govitecan-hziy versus 2.5 months on chemo; in Trop-2-medium patients, the median progression-free survival was 5.6 months versus 2.2 months, respectively; and in the Trop-2-low subgroup, these figures were 2.7 months versus 1.6 months, respectively.
Because very few patients enrolled in the trial had Trop-2 low expression, Hurvitz cautioned that these data should be interpreted with caution.
Overall survival outcomes according to Trop-2 expression levels mirrored the trend seen with progression-free survival, and across all three subgroups overall survival was improved with sacituzumab govitecan-hziy versus chemo. Across all Trop-2 expression subgroups, overall response rates were also higher with sacituzumab govitecan-hziy than chemo, though Hurvitz pointed out that there was a significant difference in response rates in the Trop-2-high subgroup.
Among sacituzumab govitecan-hziy-treated Trop-2-high patients, 44 percent saw their tumors shrink versus 1 percent in the comparator arm. These differences in response rates were less pronounced in the subgroups of patients whose tumors had medium and low Trop-2 expression status, but sacituzumab govitecan-hziy still outperformed chemo.
The treatment-related adverse events were also similar across the three Trop-2 subgroups, Hurvitz said, noting that "Trop-2 did not affect toxicity."
Germline BRCA1/2 mutations, on the other hand, are far less common than Trop-2 expression in this patient population, only tending to occur in about 10 percent of TNBC patients. Hurvitz explained that because BRCA1/2 mutations disrupt homologous recombination repair mechanisms within cancer cells, limiting their ability to repair DNA damage, the presence of these mutations could work synergistically with sacituzumab govitecan-hziy to more effectively kill TNBC tumors through synthetic lethality.
Roughly 60 percent of patients in both treatment arms had been tested at baseline for germline BRCA1/2 mutations. Only about seven percent or eight percent of patients in both treatment arms were confirmed to have BRCA1 or BRCA2 mutations.
Across all of the efficacy endpoints assessed — overall response rates, median progression-free survival, and median overall survival — both BRCA1/2-mutated and BRCA1/2-wildtype patients benefitted more from treatment with sacituzumab govitecan-hziy than chemo.
Though it wasn't pronounced enough to be considered statistically significant, Hurvitz noted that efficacy outcomes were numerically higher for sacituzumab govitecan-hziy versus chemo in the BRCA1/2-positive patient subgroup.
However, as with the exploratory Trop-2 expression analysis, Hurvitz cautioned that the data on BRCA1/2-positive patients "should be interpreted with caution," given the small sample size.
While BRCA1/2 mutation is by now a well-established predictive biomarker in various cancer indications treated with PARP inhibitors, Trop-2 is perhaps less established. After her presentation at SABCS, Hurvitz fielded several questions about Trop-2 expression and its clinical utility.
The biomarker, she said, is certainly prognostic. "Lower Trop-2 expression, if you look at the [survival] curves, is associated with worse outcome regardless of treatment," she said. "There does appear to be better outcomes associated with higher levels of expression." The biomarker analysis showed that patient outcomes improved overall with higher Trop-2 expression, but in this study, the differential benefit with the antibody-drug conjugate versus chemo did not widen based on Trop-2 expression levels.
"I don't think Trop-2 should be used to identify patients who would or would not benefit from [sacituzumab govitecan-hziy]," she concluded. "That said, is it a prognostic marker? Potentially. But I don't see how that could help an individual patient."
Going forward, Hurvitz said that researchers should consider studies addressing whether Trop-2 expression levels change over time or change in primary versus metastatic tumors. "That's something that should be looked at in a more granular way," she said, though she recognized that Trop-2 is not like PD-L1, which is not expressed in most patients and may be differentially expressed based on the tissue type and timepoint in treatment.
A ripe opportunity for additional biomarker analysis may arise in the wake of a separate ongoing Phase III study of sacituzumab govitecan-hziy, dubbed Tropics-2, evaluating the agent in hormone receptor-positive breast cancer. "There will also be a lot of tissue coming out of that study that we can sort of query for some of these questions," she said.