Editors: Donna E. Hansel, MD, PhD, chief, Division of Anatomic Pathology, and professor, Department of Pathology, University of California, San Diego; John A. Thorson, MD, PhD, associate professor of pathology, director of the Clinical Genomics Laboratory, Center for Advanced Laboratory Medicine, UCSD; Sarah S. Murray, PhD, professor, Department of Pathology, and director of genomic technologies, Center for Advanced Laboratory Medicine, UCSD; and James Solomon, MD, PhD, resident, Department of Pathology, UCSD.
Concordance between liquid biopsy and patient-matched tumor molecular testing
The use of sequence analyses of extended panels of genes to identify therapeutic targets in cancer is becoming commonplace. These assays typically rely on the availability of tissue biopsies as a source of genomic material, which can become a limitation in situations where insufficient tissue is available or an invasive procedure to collect tissue is impractical. A potential solution to this dilemma is the use of a blood-based, or liquid biopsy, approach, in which a peripheral blood sample is used as a source of tumor-derived genomic material, either in the form of cell-free DNA (cfDNA) or via circulating tumor cells (ctcDNA). Although this approach is promising, the degree to which data obtained from the material derived from blood accurately reflect the tumor genome has not been clearly determined. The authors suggest that a cautious approach to the use of liquid biopsies may be warranted and that many factors may have an impact on the comparability of findings between a tissue and liquid biopsy. They studied a group of 34 patients with pancreatic ductal adenocarcinoma (PDA) who were enrolled in a molecular registry profiling trial. All of the patients had undergone blood-based genomic profiling via commercially available assays. Twenty-three of the patients had also undergone tumor-based testing, and this group was used to evaluate the concordance between blood-based and tissue-based testing. Although the total number of mutations found per patient was similar after normalization for assay content, the authors observed significant differences in detection rates for specific gene mutations. Only a 39 percent concordance for KRAS mutation status was found, with nine of 23 patients having the same result by both assays—14 KRAS variants in tumor tissue were not detected in blood. All KRAS mutations detected in blood samples were also detected in the tumor tissue samples. A similar result was observed for TP53 mutations, with only a 26 percent concordance between tumor- and blood-based assays, with 15 TP53 variants detected in tissue but not blood. The authors also noted that a number of variant types found in tumor tissue could not be detected by the blood-based assays, as these assays were not validated for all variant types. This included deletions, indels, and splice variants. Although technical issues may have accounted for some of the discordant results, it appears that biological issues, such as lack of mutations in blood, may also have played a role. Overall, the sensitivity of the cfDNA assay, defined as the percentage of tumor tissue mutations also found in blood, was 25 percent. This is a significantly lower value than has been reported in previous studies, and the authors suggest that several variables may have contributed to these findings. First, the sample size was small and there were technical challenges with some of the blood samples—that is, low cfDNA yield. Second, the patient mix was perhaps more reflective of a real-world experience in that there was minimal preselection and many of the patients were heavily pretreated prior to the blood-based analysis. This may have had an impact on the levels of cfDNA and, therefore, variant detection. These results, in the view of this CAP TODAY contributing editor (J. T.), are somewhat disappointing because of their lack of concordance, but they serve to highlight the difficulties and potential pitfalls that may be encountered as cutting-edge approaches to cancer diagnosis and management are applied more broadly. Future study designs will benefit from the issues uncovered by this work.
Pishvaian MJ, Bender RJ, Matrisian LM, et al. A pilot study evaluating concordance between blood-based and patient-matched tumor molecular testing within pancreatic cancer patients participating in the Know Your Tumor (KYT) initiative [published online ahead of print November 8, 2016]. Oncotarget. doi:10.18632/oncotarget.13225.
Correspondence: Jonathan R. Brody at jonathan.brody@jefferson.edu
[hr]
Genomic analysis of radiotherapy response shows somatic mutations in ATM
The practice of personalized medicine, whereby genomic analyses of tumor tissue are used to inform therapeutic decisions, is based largely on growing understanding of the molecular determinants of clinical responses to drug therapies. Nucleotide or copy number variants revealed by the results of sequencing or microarray-based assays are used to guide drug therapy selection or the clinical trial of a drug that specifically targets the identified genomic lesion or a pathway affected by the lesion. While this application of genomic data to patient care is proving valuable in defining and refining the most efficacious use of drugs for a variety of cancers, it has not had a significant impact on the identification of patients most likely to benefit from radiation therapy. The authors suggest, however, that genomic analyses may have an impact on predictions of responses to radiotherapy. They initially noted a case of exceptional response to palliative radiotherapy in a 96-year-old patient with invasive squamous cell carcinoma of the tongue. Following a course of cetuximab treatment, the patient’s disease progressed within seven months. She was then treated with a standard radiotherapy regimen, which resulted in complete metabolic remission as assessed by positron emission tomography/computed tomography (PET/CT) three months post-radiotherapy. Follow-up evaluations extending to 34 months post-radiotherapy continued to show no evidence of recurrent disease. Targeted sequencing of the patient’s tumor revealed a frameshift mutation in the ATM gene and 29 other somatic mutations. Due to strong evidence linking ATM to radiation sensitivity, the authors searched their institutional database to find other patients with truncating mutations in ATM. They identified 45 additional patients, harboring a variety of primary malignancies, who had frameshift or truncating ATM mutations. Seventeen of the patients had received radiotherapy; 10 of those 17 had a minimum of three months follow-up, two of whom were excluded due to poor imaging results or multiple malignancies. The remaining eight patients demonstrated long-term disease control post-radiotherapy. Because DNA double-strand breaks (DSB) are thought to be the most significant type of lesion induced by radiotherapy and the ATM gene product is an important component of the repair response to DSB, the authors subsequently investigated the frequency of mutations affecting other genes involved in the repair of DSB. Using data from The Cancer Genome Atlas, they assessed, across 24 cancer types, the mutation frequencies for a group of 22 genes, including ATM, that are involved in the nonhomologous end joining (NHEJ) repair pathway. Of 9,064 tumor samples, 15.9 percent were found to harbor at least one genetic alteration (germline or somatic) in one of the 22 NHEJ pathway genes. Of the 22 genes surveyed, ATM was found to be the most frequently mutated, with 19.6 percent of the tumors having an NHEJ pathway alteration showing some type of ATM mutation. Although a connection between germline mutations in ATM and sensitivity to ionizing radiation has been noted before, the authors reported that this study represents the first report of the potential significance of somatic mutations in NHEJ pathway genes in predicting response to radiation therapy. While additional studies are necessary to validate these findings, the significant number of tumors harboring NHEJ mutations suggests that application of this data in a fashion similar to the current approaches for drug therapy may yield significant clinical benefits.
Ma J, Setton J, Morris L, et al. Genomic analysis of exceptional responders to radiotherapy reveals somatic mutations in ATM [published online ahead of print December 31, 2016]. Oncotarget. doi:10.18632/oncotarget.14400.
Correspondence: Nadeem Riaz at riazn@mskcc.org or Timothy A. Chan at chant@mskcc.org