Oncologists could sequence the DNA in each patient with triple-negative breast cancer (TNBC)’s tumor to strategically pinpoint treatment options for that individual, according to study findings published in Nature Medicine.
Serena Nik-Zainal, MD, and colleagues found that an algorithm, called HRDetect, classified 59% of 254 patients with having homologous recombination repair deficiency (HRD), and 67% had germline/somatic mutations of BRCA1/BRCA2, BRCA1 promoter hypermethylation, RAD51C hypermethylation, or biallelic loss of PALB2.
Across the 3-tier scoring system (high-, intermediate-, and low-risk for HRD), those considered HRDetect-high achieved better outcomes on adjuvant therapy for invasive (HR, 0.42; 95% CI, 0.20-0.87) and distant (HR, 0.31; 95% CI, 0.13-0.76) disease-free survival (DFS), compared with HRDetect-low patients.
Meanwhile, those consider HRDetect-intermediate demonstrated the poorest outcomes.
“New treatment options need to be considered for now-discernable HRDetect-intermediate and HRDetect-low categories (of patients with TNBC),” the researchers concluded. “This population-based study advocates for (whole-genome sequencing) of TNBC to better inform trial stratification and improve clinical-decisino making.”
Nik-Zainal, of the Cambridge Biomedical Research Campus in the United Kingdom, spoke with CancerNetwork® on the findings and how they could affect the evaluation of treatment in TNBC moving forward.
CancerNetwork®: What is the cost currently of whole genome sequencing like this, and is it prohibitive in individual cases at this point)?
Nik-Zainal: The cost of (whole genome sequencing) used to be prohibitive, but in truth, the steep decline in cost today means that sequencing a tumor and matched normal pair is cheaper than a standard CT scan of the chest, abdomen, and pelvis, something that would be ordered for many cancer patients at the point of diagnosis, in order to gauge the stage of their cancer disease.
The true (and more real concern) is the expertise of analysis and interpretation of (whole-genome sequencing). But that is why we have developed algorithms like HRDetect, which help future clinicians to interpret (whole-genome sequencing) data. These types of algorithms and readouts facilitate analysis and interpretation.
When could this be realized in TNBC clinical cases?
This should affect how we think about TNBC from now.
We know that TNBCs are difficult to treat. Some patients respond, others don’t. (Whole-genome sequencing) gives us insight into which ones are more likely to respond and not only identify the ones that don’t, but also understand why some of them don’t (and possibly direct them to alternative therapies).
But what we need to get over is the “fear” of (whole-genome sequencing). As mentioned before, the 2 main hurdles are the fear of cost and the fear of analysis/interpretation. The first 1 is rapidly becoming a myth and the second is being realized by new algorithms.
When could whole-genome sequencing be used in other cancers?
Sequencing, analysis, and interpretation can now be done really quite quickly. The tools for analysis and interpretation are tumor agnostic. It can be done in any tumor today already.
In your study, why did the intermediate-risk group fare poorest, compared with those in the low-risk group)
We don’t have an explanation for all of them. But we do have clues for some of those samples. That comes from looking at the full picture of all the drivers and the mutational signatures (patterns of mutations that arise as the cell turned from being a normal cell into a tumor cell).
In using the Illumina machine to sequence the 250-plus genomes, was this work a product of massively parallel sequencing?
Yes, it was the product of massively parallel sequencing. In a whole-genome sequencing experiment, we get to read out the entire genetic sequence of the tumor. We are not just looking at coding sequences (genes) or a handful of genes, we are looking at the entire genome. That means that we have a very complete view of the mutations that have accumulated over the time of the development of the tumor.
What are the clinical takeaways for oncologists from the study?
We used an algorithm called HRDetect to classify the tumors. The patients that fall in the category of HRDetect-high (cancers) appear to have a good outcome on current therapies for TNBC. Which is great, but perhaps their outcomes could be improved further with the application of PARP inhibitors.
The patients that fall in the category of HRDetect-intermediate and -low (cancers), on the other hand, tend to have poorer outcomes on current therapies. Some of them have other features that are potentially targetable. Perhaps new clinical trials are not required for these patients.
Staaf J, Glodzik D, Bosch A, et al. Whole-genome sequencing of triple-negative breast cancer in a population-based clinical study. Nat Med. 2019 Sep 30. doi: 10.1038/s41591-019-0582-4