Evaluation of a novel proprietary liquid biopsy assay to identify high risk breast ductal carcinoma in situ (D

Year Submitted: 2025
Published Date: 07 May 2025

Tags: Technology, Care Process ＆ Redesign, Quantitative Research

About this Content

Aims

The objective of the study is to identify independent predictors of an invasive component of DCIS and evaluate the somatic alterations in DCIS and T1aN0 breast cancer.

Background

Most women with ductal carcinoma in situ (DCIS) have an excellent prognosis and the DCIS may not even progress during the lifetime of some women. However, for women with a pre-operative diagnosis of DCIS, there is currently no means of identifying those with an invasive component or who are at high risk of invasive progression or relapse. Aims. We aim to develop a means of identifying women with DCIS likely to have an invasive component. Study objectives are to: 1) Identify factors predictive of an invasive component in women pre-operative diagnosis of DCIS, 2) Evaluate a novel proprietary NGS assay to identify high-risk DCIS.

Methods

We will prospectively recruit 120 women with a pre-operative diagnosis of DCIS and another 15 women with T1aN0M0 disease for comparison. Standard demographic, clinicopathological, radiological data and treatment details will be analysed for a correlation with an invasive upgrade. Multivariate analyses will be done to identify independent risk factors. Using a novel NGS assay we have developed, we will evaluate blood plasma samples from 70 patients (10 low-grade DCIS, 10 intermediate-grade DCIS, 40 high-risk DCIS (most likely to have an invasive grade) and 10 T1aN0M0, to serve as comparison); these being patients recruited earlier. We will examine the concordance between cell free DNA in the plasma and tumour DNA by performing the NGS assay on archival tumour sections from 8 of these patients and correlating with matched plasma samples. Somatic alterations as detected on NGS assay will be analysed together with factors found significant on the earlier analyses to identify independent predictors of invasive carcinoma in DCIS. Findings will provide important preliminary data on the performance of the NGS assay and will allow us to optimise it further if necessary. Potentially, this work will provide a means of reducing the over-treatment of DCIS.

Results

We reviewed all cases where DCIS was diagnosed on biopsy and examined the rates of upgrade to invasive cancer found at surgery. We also examined clinic-pathological factors that were associated with this upgrade risk. We collaborated with NUH, KKWCH and CGH for greater sample size. We have completed the repeated analyses with a combined dataset from TTSH, NUH and KKWCH. This study to review the incidence of invasive upgrade and its risk factors using aggregate data from the 4 hospitals will be submitted for publication. Manuscript was submitted for review and the work was revised. We have expanded the work to include the development of a predictive model that can be used during the discussion for management of DCIS. This work is expected to have a greater clinical impact than previously as it presents a simple-to-use model, using parameters that are easily available from routine clinical, radiological and pathological assessments. This is further validated in a follow-on grant (NTF_DEC2021_1_C2_CR_02). We faced problems with the NGS assay and eventually decided not to proceed with the liquid biopsy assay as the current technology is unable to detect low levels of circulating DNA that is common in DCIS. We evaluated another biomarker, NELF, that has shown promising results in another work that we are doing. NELF expression levels appear to be correlated with the grade of tumour and we want to explore the possibility of NELF being a marker of DCIS transformation to invasive cancer. This work is now being further evaluated as part of another grant (NTF_SRP_P2; Precision in Cancer Prevention). The predictive model of invasive upgrade in DCIS will be an important decision-making tool in the clinics. It potentially allows clinicians to weigh the benefit of treatment of especially DCIS detected incidentally in elderly women. With this tool, we will potentially be able to conduct studies to finally tailor the need for DCIS treatment.

Conclusion

Despite the failure to develop a liquid biopsy assay that can predict for invasive upgrade in DCIS, we were able to develop a model that can be used in the clinics to predict the likelihood of invasive upgrade. This model will be an important decision-making tool in the discussion with patients on how DCIS should be managed. In addition, we carried out preliminary work on a biomarker that could potentially also identify DCIS at high-risk of invasive upgrade or transformation. This is being further evaluated in another grant.

Lessons Learnt

Work was significantly affected by the COVID-19 outbreak. Personnel re-deployments and changes at GIS and at TTSH Pathology meant that Pathology could not cut the FFPE sections for us and GIS could not carry out the bench work as planned. We resumed after the Circuit breaker, but this was affected again during the TTSH cluster outbreak when research activities were suspended. We are just about to resume but it seems that this will now be again postponed given the new heightened alert measures. Sensitivity of the 1q21.3 assay has not been good, likely because of the low levels of ctDNA present in DCIS.

Keywords

Effectiveness, invasive component, ductal carcinoma in situ (DCIS), Oncology

Innovators' Details

Connect with this contributor!

Assoc Prof Tan Ern Yu - Ern_yu_tan@ttsh.com.sg

Project Attachment