NEW YORK (GenomeWeb) – Analyzing the cancer genomes of breast cancer metastases from patients who have succumbed to the disease may help reveal the genomic alterations that led to lethality, according to researchers from the Peter MacCallum Cancer Center in Melbourne, Australia.
The researchers sequenced the exomes of and ran SNP arrays on metastatic lesions taken from four patients who died of breast cancer as part of the community-based rapid autopsy program, CASCADE, and reported their results this week in PLOS Medicine.
Little is known about the genomic changes that happen in the most advanced stages of breast cancer, particularly in cases where the cancer leads to death. In order to analyze metastases after a patient has died, autopsies must be performed rapidly in order to preserve the tissue.
In the study, the researchers analyzed samples from three patients with estrogen-receptor positive, human epidermal growth factor receptor 2-negative breast cancer and one patient with triple negative breast cancer, who were part of the institutional prospective CASCADE trial. From each patient the researchers analyzed between five and 12 metastatic lesions taken at autopsy, along with primary tumor and longitudinal metastatic biopsies that had been taken during life.
The four cases each represented difficult clinical scenarios, including a late relapse after early stage disease, de novo metastatic disease, discordant disease response, and not responding to treatment.
The team found significant differences in the mutational burden and processes as well as in the driver mutations present in each case. Even within the same patient, they found significant heterogeneity between the different metastases as well as between the primary tumor and metastatic sites.
In ER-positive breast cancer, late relapse is a significant problem, with declining rates of survival beyond 10 years from diagnosis. One patient exemplified this issue, who relapsed seven years after her initial diagnosis. Genomic analysis of the primary and metastatic samples identified the subclone that led to the relapse. The subclone was not detectable in the primary tumor although the researchers did identify tumor driver mutations and copy number alterations before the clinical diagnosis of metastasis, which has "important implications" for the "expectation that genomic assays from a single time point are able to predict clinical outcome and guide therapy," the authors wrote. Instead, the findings highlight the potential for other ways of monitoring disease recurrence, potentially through analyzing circulating tumor DNA.
In another ER-positive case, the researchers observed a very different disease course. Rather than a prolonged remission, this patient had the "unexpected finding that the earliest subclone was able to metastasize into the ovary." In addition, this patient also had "widespread copy number derangement" present in all samples. The different samples had different copy number profiles, which also explained the patient's discordant response to treatment.
In the patient with triple negative breast cancer, the researchers identified a mutational signature that seemed to be carcinogen induced, which "raises questions about the etiology of aggressive triple negative breast cancers," the researchers wrote. The signature, dubbed signature 17, has been previously reported in multiple cancer types, although its etiology is unknown. This mutational signature was found primarily in the metastatic lesions collected at autopsy, comprising between 9 percent and 45 percent of the private mutations in the lung and liver metastases.
The researchers hypothesized that the signature may have arisen from cancer cells that survived treatment. To figure out what treatment might have been responsible, they worked backwards from when the mutational signature 17 was spotted. The only treatment that the patient received after the last biopsy to not contain the signature was a pan aurora kinase inhibitor, to which the patient did not respond. The authors wrote, however, that it is unclear how the inhibitor could have caused the mutational signature.
The researchers also identified known and one potentially novel mechanism of treatment resistance. Many of the autopsy samples had subclones with resistance mutations. For instance, one of the ER-positive patient had three different known mutations to the ESR1 gene in different subclones, which confers resistance to endocrine therapy. Identifying these mutations is important, the authors noted, because they can be treated with alternative endocrine therapy.
The team also found evidence that different subclones may respond differently to treatment and may impact evolution in different ways. For instance, "resistance to chemotherapy agents may arise from augmentation of existing oncogenic or tumor suppressor signaling pathways rather than direct resistance through altered drug targets or drug metabolism," the authors wrote.
Overall, the study, although limited in sample size, demonstrates the importance of comprehensively analyzing the genomic profile of not just patients' primary tumors, but of all the metastatic lesions, as well, in order to understand disease progression and evolution.
"Our prospective rapid autopsy program, which continues to accrue in breast cancer and other cancer types, as well as other international efforts, will be essential to help us understand how cancer disseminates and ultimately becomes resistant to treatment," the authors wrote.