Abstract P5-01-24: DNA palindromes as novel genomic targets for unbiased, genome-wide, and rapid pan-cancer screening

Abstract

Abstract Cancer continues to have a significant medical, economic, and psychological burden on patients worldwide. One strategy for reducing this impact involves the early detection of cancer biomarkers. Structural chromosomal aberrations called DNA palindromes have immense potential to be used as early markers in tumorigenesis. Palindromic sequences in the genome are structural genetic phenomena where a sequence of nucleotides is identical to its reverse compliment. Various independent studies have identified this distinctive pattern of genome instability in a wide range of cancers including breast, pancreatic, and ovarian cancers. We have shown previously that DNA palindromes underlie the amplification of ERBB2 in breast cancer. This suggests that DNA palindromes are the candidate breast cancer biomarker. A convenient and sensitive approach for identifying palindromes in breast tumor DNA could be a promising diagnostic test. Palindrome detection from tumor DNA in blood (liquid biopsy) could circumvent the need for invasive biopsies. Here, we develop an unbiased genome-wide technique for detecting DNA palindromes as a genomic target for breast cancer screening in the clinic. Using genome-wide analysis of palindrome formation (GAPF), we are able to enrich for DNA palindromes from samples of genomic DNA isolated from cells and tumor tissues. Following a denaturation step, DNA palindromes naturally form double-stranded DNA with itself because these sequences are physically tethered together and can form energetically favorable base pairs. Non-palindromic DNA remains single-stranded and subsequently removed by single-strand specific nucleases thereby dramatically improving the signal-to-noise ratio. This approach confers advantages over other methods because it simultaneously (1) amplifies target signal and (2) reduces background noise without employing sub-sampling techniques. We first used a cancer cell line Colo320DM containing a known palindromic junction and normal fibroblasts IMR90. By integrating GAPF with next generation sequencing (GAPF-Seq), we mapped sequenced reads to a reference genome and developed an algorithm to demarcate DNA palindromes throughout the genome. We are able to identify the known DNA palindrome in Colo320DM in a 10-fold dilution in IMR90 DNA. We then used GAPF-Seq to identify tumor-specific DNA palindromes in matched pairs of tumor and normal breast cancer tissue. From 100 ng of input DNA, we generated a profile of more than 300 tumor-specific DNA palindromes that were not found in the matched normal tissue. To optimize GAPF for use in cancer screening, we expect to detect DNA palindromes in the presence of 100-fold excess of non-palindromic DNA from as little as 10 ng of input DNA. We will then use matched plasma and buffy coat samples to generate whole-genome palindromic profiles from liquid biopsies. The data can also be used for tracing clonal evolution by comparing individual bins in the palindrome profiles of tumor tissue and plasma cell-free DNA from the same patient. Finally, we will optimize GAPF-Seq for minimal computational processing and data storage requirements. Due to the amplification of palindromes and reduction of background signal, GAPF-Seq may be able to identify tumor-specific breast cancer markers from as few as 1 million reads and 250 MB of data, which would be ideal for wide use in screening for breast tumors in the general population and monitoring disease outcomes with multiple blood samplings. We believe GAPF-Seq addresses a need to develop an unbiased, genome-wide, and rapid approach for breast cancer screening. Citation Format: Michael M. Murata, Lila Mouakkad, Dolores Di Vizio, Armando E. Giuliano, Hisashi Tanaka. DNA palindromes as novel genomic targets for unbiased, genome-wide, and rapid pan-cancer screening [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P5-01-24.