Blog | Feb. 15, 2023
With liquid biopsy, sensitivity matters
In Part I, we covered the high tissue biopsy to liquid biopsy concordance across six different colorectal cancer studies (available here) with an Overall Percent Agreement (OPA) calculated across 913 clinical samples to be 90.3%. Here we review tissue and liquid biopsy concordance studies in non-small cell lung cancer (NSCLC), breast cancer, and melanoma.
Both activating and T790M resistance mutations in non-small cell lung cancer (NSCLC) were examined in a dual observational and Phase I trial of an EGFR inhibitor rociletinib in Karlovich et al. (2016 Clinical Cancer Res.)1 Tissue was tested using Standard-of-Care (SoC) real-time PCR, and of 63 patients tested the OPA of EGFR activating mutations between tissue and the OncoBEAM EGFR test on plasma was 81%.
Looking at the same 63 patients for the EGFR T790M resistance mutation, the OPA between tissue and plasma was 67%; however, an additional 9 patients detected EGFR T790M in plasma that went undetected in tumor tissue, and another 9 patients detected EGFR T790M in plasma where insufficient tissue was available for analysis. Overall the OncoBEAM test identified more T790M-positive patients (51) than did the tumor test (45).
Another study also examined EGFR mutations in NSCLC for a Phase I trial (“AURA”) of an EGFR inhibitor AZD9291 (approved as TAGRISSO® or osimertinib) in Thress et al. (2015 Lung Cancer)2. They used a cross-platform comparison across two real-time PCR platforms for EGFR (cobas EGFR Mutation Test and therascreen EGFR amplification refractory mutation assay) and two digital PCR platforms (Bio-Rad Droplet Digital™ PCR and OncoBEAM BEAMing digital PCR). Their preliminary assessment across 38 samples achieved 95% concordance between tissue testing and OncoBEAM for Exon 19 deletions and L858R mutations, and 70% concordance between tissue testing and OncoBEAM for T790M.
The authors note that the concordance between plasma testing with the cobas EGFR Mutation Test and OncoBEAM (instead of tissue against OncoBEAM and plasma) was over 90%, with 14/20 of the discordant tissue versus plasma results were in perfect agreement when comparing the two orthogonal plasma testing technologies; the remaining 6/20 discordant cobas plasma versus OncoBEAM plasma results all had MAF of <0.2%, below the detection limit of the cobas EGFR Mutation Test, and likely due to tumor heterogeneity to explain the tissue versus plasma discordance.
In a study examining third-generation tyrosine kinase inhibitor (TKI) activity against EGFR T790M resistance, Oxnard et al. (2016 J. Clinical Oncol)3 in their analysis determined Objective Response Rate (ORR) and Progression-Free Survival (PFS) in T790M-positive and T790M-negative patients. Their results concluded ORR and median PFS were similar in T790M-positive plasma and T70M-positive tumor; thus with a validated assay some patients could avoid a tumor biopsy for T790M tissue testing and use a plasma-based test instead.
In Rudolph et al. (2016 BMC Cancer)4, more than 600 clinical breast cancer samples were tested for a specific AKT1 mutation [G49A:E17K] and overall its mutation prevalence was 6.3%. All of the tissue samples via OncoBEAM for mutations in both the ATK1 and PIK3CA genes; a subset of 90 samples were also tested via a broad-panel NGS test.
Within this sample set, there were 179 breast cancer tissue samples that were AKT1E17K mutation-positive. Overall plasma concordance with OncoBEAM yielded an OPA of 81.6%. Of 121 metastatic breast cancer samples that were PIK3CA mutation-positive in tissue, the plasma concordance had an OPA of 86.8%.
In Higgins et al. (2012 Clin Cancer Res)5 49 retrospective tumor and temporally-matched plasma samples were compared for PIK3CA mutations, and achieved an OPA of 100%. With an additional 41 prospective cohort tumor samples with matched tissue and plasma another OPA of 100% was observed.
In Di Leo et al. (2018 Lancet Oncol.)6 a group of postmenopausal women with hormone-receptor positive, HER2-negative advanced breast cancer participated in an mTOR inhibition Phase III trial (called BELLE-3). ctDNA PIK3CA status was determined by OncoBEAM, while tissue was evaluated by real-time PCR. Of the 256 samples evaluated, an overall concordance of 83% was reported, with 80% sensitivity and 87% specificity.
Using the OncoBEAM BRAF and NRAS assay, Rowe et al. (2018 Molecular Oncol.)7 determined the BRAF and NRAS tissue and plasma mutation status across 55 patient samples collected prospectively. This work investigated the clinical utility (i.e. impact on clinical outcomes and interpretation of radiographic images) of measuring ctDNA in patients with metastatic or high-risk resected melanoma.
Both tissue and plasma were tested with OncoBEAM BRAF and NRAS assay technology, and across the 55 patient samples the researchers determined an OPA of 90.9%.
In another prospective study Haselmann et al. (2018 Clinical Chem.)8 examined BRAF mutation status in both tissue and plasma to correlate with the clinical course of disease and with response to treatment. Of 187 patient tissue samples tested (all were Stage I or Stage II as part of ‘Study 1’ in their report), concordance between tissue testing by OncoBEAM compared to plasma testing achieved an OPA of 90.9%.
The correlation to the clinical course of disease and response to treatment across many more patients (n=1204, “Study 2 follow-up of patients”) achieved an OPA of 95.7%.
If you would like to access our concordance data for colorectal cancer across tissue and plasma samples analyzed by OncoBEAM, you can find it here.
For more information about how OncoBEAM enhanced digital PCR can help in your liquid biopsy studies, please contact us here.