Automated capture-based NGS workflow: one thousand patients experience in a clinical routine framework.

OBJECTIVES: The Next Generation Sequencing (NGS) based mutational study of hereditary cancer genes is crucial to design tailored prevention strategies in subjects with different hereditary cancer risk. The ease of amplicon-based NGS library construction protocols contrasts with the greater uniformity of enrichment provided by capture-based protocols and so with greater chances for detecting larger genomic rearrangements and copy-number variations. Capture-based protocols, however, are characterized by a higher level of complexity of sample handling, extremely susceptible to human bias. Robotics platforms may definitely help dealing with these limits, reducing hands-on time, limiting random errors and guaranteeing process standardization. METHODS: We implemented the automation of the CE-IVD SOPHiA Hereditary Cancer Solution™ (HCS) libraries preparation workflow by SOPHiA GENETICS on the Hamilton's STARlet platform. We present the comparison of results between this automated approach, used for more than 1,000 DNA patients' samples, and the performances of the manual protocol evaluated by SOPHiA GENETICS onto 240 samples summarized in their HCS evaluation study. RESULTS: We demonstrate that this automated workflow achieved the same expected goals of manual setup in terms of coverages and reads uniformity, with extremely lower standard deviations among samples considering the sequencing reads mapped onto the regions of interest. CONCLUSIONS: This automated solution offers same reliable and affordable NGS data, but with the essential advantages of a flexible, automated and integrated framework, minimizing possible human errors and depicting a laboratory's walk-away scenario.

Characterization of New ATM Deletion Associated with Hereditary Breast Cancer.

Next-generation sequencing (NGS)-based cancer risk screening with multigene panels has become the most successful method for programming cancer prevention strategies. ATM germ-line heterozygosity has been described to increase tumor susceptibility. In particular, families carrying heterozygous germ-line variants of ATM gene have a 5- to 9-fold risk of developing breast cancer. Recent studies identified ATM as the second most mutated gene after CHEK2 in BRCA-negative patients. Nowadays, more than 170 missense variants and several truncating mutations have been identified in ATM gene. Here, we present the molecular characterization of a new ATM deletion, identified thanks to the CNV algorithm implemented in the NGS analysis pipeline. An automated workflow implementing the SOPHiA Genetics' Hereditary Cancer Solution (HCS) protocol was used to generate NGS libraries that were sequenced on Illumina MiSeq Platform. NGS data analysis allowed us to identify a new inactivating deletion of exons 19-27 of ATM gene. The deletion was characterized both at the DNA and RNA level.

Hereditary Pancreatic Cancer: A Retrospective Single-Center Study of 5143 Italian Families with History of BRCA-Related Malignancies.

The identification of BRCA mutations plays a crucial role in the management of hereditary cancer prevention and treatment. Nonetheless, BRCA-testing in pancreatic cancer (PC) patients is not universally introduced in clinical practice. A retrospective analysis was conducted, firstly, to evaluate the rate of BRCA-positive families among those presenting a family history of PC besides breast and/or ovarian cancer. Secondly, the relationship between BRCA pathogenic variants and PC risk was evaluated. Finally, the characteristics of PC developed in BRCA families were described. Among 5143 family trees reporting breast and/or ovarian cancer cases, 392 showed a family history of PC. A total of 35 families (24.5% selected by the Modena Criteria and 21.3% by the NCCN Criteria) were positive to BRCA testing. Among the BRCA1 mutations, 36.8% were found within a region defined by c.3239⁻c.3917, whilst 43.7% of BRCA2 mutations were located within c.7180⁻c.8248. This study confirmed that an increase in the rate of positive tests in families with PC when associated to breast and/or ovarian tumors. Moreover, this analysis indicated two possible Pancreatic Cancer Cluster Regions that should be verified in future research. Finally, PC in families with breast and/or ovarian cancer history, particularly in BRCA families, were diagnosed at younger age and showed better one-year overall survival.

Genomic alterations at the basis of treatment resistance in metastatic breast cancer: clinical applications.

The standard of care for breast cancer has gradually evolved from empirical treatments based on clinical-pathological characteristics to the use of targeted approaches based on the molecular profile of the tumor. Consequently, an increasing number of molecularly targeted drugs have been developed. These drugs target specific alterations, called driver mutations, which confer a survival advantage to cancer cells. To date, the main challenge remains the identification of predictive biomarkers for the selection of the optimal treatment. On this basis, we evaluated a panel of 25 genes involved in the mechanisms of targeted treatment resistance, in 16 primary breast cancers and their matched recurrences, developed during treatment. Overall, we found a detection rate of mutations higher than that described in the literature. In particular, the most frequently mutated genes were ERBB2 and those involved in the PI3K/AKT/mTOR and the MAPK signaling pathways. The study revealed substantial discordances between primary tumors and metastases, stressing the need for analysis of metastatic tissues at recurrence. We observed that 85.7% of patients with an early-stage or locally advanced primary tumor showed at least one mutation in the primary tumor. This finding could explain the subsequent relapse and might therefore justify more targeted adjuvant treatments. Finally, the mutations detected in 50% of relapsed tissues could have guided subsequent treatment choices in a different way. This study demonstrates that mutation events may be present at diagnosis or arise during cancer treatment. As a result, profiling primary and metastatic tumor tissues may be a major step in defining optimal treatments.

Impact of mutational status on outcomes in myelofibrosis patients treated with ruxolitinib in the COMFORT-II study.

The JAK1/JAK2 inhibitor ruxolitinib produced significant reductions in splenomegaly and symptomatic burden and improved survival in patients with myelofibrosis (MF), irrespective of their JAK2 mutation status, in 2 phase III studies against placebo (COMFORT-I) and best available therapy (COMFORT-II). We performed a comprehensive mutation analysis to evaluate the impact of 14 MF-associated mutations on clinical outcomes in 166 patients included in COMFORT-II. We found that responses in splenomegaly and symptoms, as well as the risk of developing ruxolitinib-associated anemia and thrombocytopenia, occurred at similar frequencies across different mutation profiles. Ruxolitinib improved survival independent of mutation profile and reduced the risk of death in patients harboring a set of prognostically detrimental mutations (ASXL1, EZH2, SRSF2, IDH1/2) with an hazard ratio of 0.57 (95% confidence interval: 0.30-1.08) vs best available therapy. These data indicate that clinical efficacy and survival improvement may occur across different molecular subsets of patients with MF treated with ruxolitinib.

Deep sequencing unearths nuclear mitochondrial sequences under Leber's hereditary optic neuropathy-associated false heteroplasmic mitochondrial DNA variants.

Leber's hereditary optic neuropathy (LHON) is associated with mitochondrial DNA (mtDNA) ND mutations that are mostly homoplasmic. However, these mutations are not sufficient to explain the peculiar features of penetrance and the tissue-specific expression of the disease and are believed to be causative in association with unknown environmental or other genetic factors. Discerning between clear-cut pathogenetic variants, such as those that appear to be heteroplasmic, and less penetrant variants, such as the homoplasmic, remains a challenging issue that we have addressed here using next-generation sequencing approach. We set up a protocol to quantify MTND5 heteroplasmy levels in a family in which the proband manifests a LHON phenotype. Furthermore, to study this mtDNA haplotype, we applied the cybridization protocol. The results demonstrate that the mutations are mostly homoplasmic, whereas the suspected heteroplasmic feature of the observed mutations is due to the co-amplification of Nuclear mitochondrial Sequences.