NEW YORK – A new genomic, immune profiling, and imaging study suggests there may multiple signatures for predicting response to a PARP inhibitor and checkpoint immunotherapy drug combination in advanced ovarian cancer patients.
Researchers from the US and Finland used targeted gene panel sequencing, targeted methylation profiling, copy number analyses, tumor mutational burden analyses, and other approaches to characterize genomic features in tumor samples from 62 recurrent ovarian cancer patients enrolled in a Phase I/II trial of the PD-1-targeting drug pembrolizumab (Merck's Keytruda) and the PARP inhibitor niraparib (GlaxoSmithKline's Zejula).
Together with multiplexed single-cell imaging data, these immunogenomic profiles led the team to two potential prognostic signatures: a homologous recombination DNA repair-related mutational signature and an immune score reflecting the tumor microenvironment. The findings appeared in Nature Communications on Thursday.
"By taking these factors into account, researchers leading trials of this combination in patients with advanced, chemotherapy-resistant ovarian cancer may select individuals who may respond to this combination of drugs," co-senior author Panagiotis Konstantinopoulos, director of translational research at the Dana-Farber Cancer Institute's gynecologic oncology department, said in a statement.
Conversely, Konstantinopoulos explained, such proposed markers may eventually "help to ensure that patients for whom a PARP inhibitor-checkpoint inhibitor combination won't be beneficial can focus on other clinical trials of treatments that may be more effective for them."
Long-term responses to immune checkpoint immunotherapy alone have been documented in a relatively small subset of ovarian cancer cases, the team noted, prompting interest in combined treatment strategies that include PARP inhibitors — drugs used to treat ovarian cancer cases marked by risky BRCA1 or BRCA2 mutations.
As part of the TOPACIO clinical trial, the researchers focused on formalin-fixed, paraffin-embedded tumor samples collected from 62 individuals with niraparib plus pembrolizumab-treated recurrent ovarian cancer patients, who had completed between one and five lines of prior therapy or three previous treatments, on average.
In this patient group, the team noted that the PARP and PD-1 inhibitor combination had an overall response rate of 18 percent, including three patients with a complete response and eight patients with partial responses.
Moreover, the authors noted that 65 percent of patients showed some clinical benefit, "clearly exceeding the expected activity of niraparib or pembrolizumab as monotherapies in recurrent platinum-resistant ovarian cancer."
Even so, some of the advanced ovarian cancer patients had stronger responses to niraparib and pembrolizumab than others — treatment response differences that the team explored in more detail using targeted sequencing panels centered on cancer-related genes or genes involved specifically in DNA repair, along with BRCA1 and RAD51C methylation profiling, and other genomic analyses.
BRCA1/2 mutation status, homologous recombination status, or PD-L1 status did not appear to provide clear insights into which patients were most likely to respond to the combined treatment, despite the documented insights that such markers can provide for patients receiving PARP inhibitors or immune checkpoint inhibitor alone.
However, the researchers did see that ovarian cancer patients with a set of tumor markers known as mutational signature 3, involving defects in homologous recombination-based DNA repair mechanisms, appeared to benefit from the PARP-PD-1 inhibitor combo.
The team saw mutational signature 3 in just over half of the 39 patients assessed using a targeted sequencing panel, which tested 447 genes in the tumor.
With parallel immune microenvironment expression, immune scoring, and single-cell antigen imaging approaches, the researchers unearthed combination treatment response-related immune markers related to exhausted CD8-positive T cells in the tumor microenvironment — an apparent prognostic immune signature that they delved into in more detail using spatial insights from the single-cell experiments.
The investigators noted that the "[p]resence of one of both features [mutational signature 3 and/or immune features in the microenvironment] associates with an improved outcome."
In contrast, they reported, advanced ovarian cancer cases with tumors lacking both signatures appeared far less prone to respond to the PARP inhibitor and checkpoint immunotherapy inhibitor drug combination.
"Our study highlights that careful analysis of genomic information and single-cell spatially resolved data from clinical samples can provide valuable information on the determinants of response to therapy," the authors wrote, "and accelerate the development of predictive biomarkers to aid in patient stratification."
They noted that biomarkers for combined PARP inhibitor-checkpoint blockade response "is particularly relevant for patients with platinum-resistant ovarian cancer who have poor prognosis and, therefore, require careful selection of their next treatment regimen."