CHICAGO – In an exploratory analysis of the Phase II ROME trial, combining liquid and tissue biopsy results to guide molecularly targeted therapy increased detection of actionable mutations and improved overall and progression-free survival for patients compared to the standard approach.
"Integrating liquid and solid biopsy offers a comprehensive approach to tumor characterization," Paolo Marchetti, professor of medical oncology at Sapienza University of Rome, said in his presentation of the results at the American Association for Cancer Research's annual meeting Tuesday.
In the ROME trial, investigators with Italy's Foundation for Personalized Medicine are evaluating the feasibility, efficacy, and safety of molecular tumor board (MTB)-directed targeted therapies and immunotherapies compared to the standard of care in patients with advanced solid tumors. To enroll in the trial, patients were screened for actionable alterations by Foundation Medicine's tissue-based FoundationOne CDx and its blood-based FoundationOne Liquid CDx.
FoundationOne CDx is a next-generation sequencing test that detects substitutions, insertion and deletion alterations, and copy number alterations (CNAs) in 324 genes and select gene rearrangements, as well as genomic signatures including microsatellite instability and tumor mutational burden, using DNA isolated from formalin-fixed, paraffin-embedded tumor tissue specimens. FoundationOne Liquid CDx similarly surveys 311 genes, rearrangements in four genes, and CNAs in three genes using circulating cell-free DNA isolated from plasma derived from anti-coagulated peripheral whole blood of cancer patients.
These patients were then randomized to the MTB arm or standard-of-care arm. In the MTB arm, a board of precision oncology experts determined the treatment patients received based on their tumor biomarkers, regardless of histology, while in the standard-of-care arm, doctors chose the standard treatment to give based on histology in line with the Italian Association of Medical Oncology guidelines. Nearly 1,800 patients were screened for the trial, and 400 patients were randomized to standard or tailored treatment.
In September 2024, Marchetti's collaborator Andrea Botticelli, a professor of oncology at Sapienza University of Rome, presented results from the trial showing that patients responded better to MTB-directed therapy than to standard therapies. The objective response rate in the targeted therapy group was 17 percent versus 9.5 percent for patients on standard therapy. Among patients who received an MTB-guided therapy, median progression-free survival was 3.7 months, compared to 2.8 months on standard-of-care treatment.
In this new analysis, the researchers wanted to explore how concordance and discordance in detecting actionable alterations between liquid and tissue biopsies impacted patient outcomes in the trial. Concordance was when both types of tests detected the same significant alterations. Discordance meant only one of the tests picked up significant alterations.
Overall, the rate of concordance between biopsies was 49 percent. Among the 203 patients with discordant biopsies, actionable mutations were detected exclusively in tissue biopsies in 35 percent of patients and only in liquid biopsies in 16 percent of patients. Test failures were the culprit in 21 percent of discordant biopsies. Another 35 percent of test disagreements were attributable to discordant high tumor mutational burden detection, 1 percent to microsatellite instability discrepancies, and 43.3 percent to differences in detection of molecular alterations. Marchetti said the PI3K/PTEN/AKT/mTOR and ERBB2 pathways had the highest discordance rates.
For patients who had concordant tissue and liquid biopsies, median progression-free survival was 4.93 months on targeted therapy and 2.80 months on standard therapy. But for patients with discordant biopsies, median progression-free survival was three months in patients whose targeted therapy was guided by a tissue biopsy only and two months for patients whose treatment was based on liquid biopsy only.
Marchetti concluded that the substantial increase in detection of actionable alterations and the significant improvement of survival outcomes with combined tissue and liquid biopsy concordance provides strong support for the importance of integrating both modalities in precision oncology approaches.
Marchetti cautioned that this is an exploratory analysis without predefined statistical power for subgroup comparisons, which limits the generalizability of the findings. "[Tissue and liquid biopsy] concordance as a predictive biomarker still requires prospective validation," Marchetti said, adding that the study did not evaluate alternate sites for liquid biopsy sampling such as pleural effusion or cerebrospinal fluid.