No abstract

The aim of this study was to determine the frequency of EGFR, KRAS, BRAF, and HER-2 mutations in brain metastases from non-small cell lung carcinomas (BM-NSCLC). A total of 77 samples of BM-NSCLC were included and 19 samples of BM from breast, kidney, and colorectal tumors were also studied as controls. These samples were collected from patients followed between 2008 and 2011 at Poitiers and Nice University Hospitals in France. The frequencies of EGFR, KRAS, BRAF, and HER-2 mutations in BM-NSCLC were 2.6, 38.5, 0, and 0% respectively. The incidence of KRAS mutation was significantly higher in female and younger patients (P < 0.05). No mutations of the four genes were found in BM from breast or kidney. However, among six BM from colorectal tumors, we identified KRAS mutations in three cases and BRAF mutations in two other cases. This study is the largest analysis on genetic alterations in BM-NSCLC performed to date. Our results suggest a low frequency of EGFR mutations in BM-NSCLC whereas KRAS mutations are as frequent in BM-NSCLC as in primitive NSCLC. These results raise the question of the variability of the brain metastatic potential of NSCLC cells in relation to the mutation pattern.

IMPORTANCE Cetuximab was recently proposed for advanced cutaneous squamous cell carcinomas (cSCC); however, its efficacy is inconsistent and identification of predictive biomarkers for response is necessary.OBJECTIVE To search for somatic mutations of the HRAS, KRAS, NRAS, BRAF, and EGFR genes in patients with advanced cSCC treated with cetuximab; and to investigate the efficacy and tolerance of cetuximab according to these mutations. DESIGN, SETTING, AND PARTICIPANTSA multicentric and retrospective study of 31 patients (22 men, 9 women) with histologically confirmed advanced cSCC carried out in 1 department of dermatology and 2 departments of medical oncology in France between January 2008 and December 2014. The median age of participants was 86 years (range, 48-96 years).INTERVENTIONS Mutational status was determined by pyrosequencing method, allelic discrimination, or Sanger sequencing. Patients were treated by single-agent cetuximab. MAIN OUTCOMES AND MEASURESThe primary end point was the incidence of somatic mutations of the RAS, BRAF, and EGFR genes and association of cetuximab efficacy with these mutations was investigated by using Fisher test. Secondary end points were the disease control rate (DCR) at week 6, the progression free-survival (PFS), overall survival (OS), and safety profile of cetuximab.RESULTS Thirty-one samples of cSCC from 31 patients were analyzed. Only 2 RAS mutated samples (6.5%) were identified. The first harbored a NRAS point mutation (c.35G>A) in codon 12, resulting in a p.G12D substitution. The second sample presented a HRAS point mutation (c.38G>T) in codon 13, resulting in a p.G13V substitution. No mutation of KRAS, BRAF, and EGFR genes at the investigated loci was found. Two patients with NRAS and HRAS mutations showed a partial and complete response to cetuximab, respectively. The mean duration of follow-up was 19 months. At week 6, the disease control rate was 67.8%. The median OS was 13 months and the median PFS was 9 months. All patients could continue cetuximab treatment without dose reduction. CONCLUSIONS AND RELEVANCEEven in elderly patients with advanced cSCC, cetuximab was efficacious and well-tolerated. This suggests that cetuximab is certainly warranted in the treatment of advanced cSCC. However, it is also important to identify tumor specific mutations that may determine response to treatment and prognosis for the disease. We have identified here that the incidence of RAS, BRAF, and EGFR mutations is low in cSCC.

Personalized medicine has gained increasing importance in clinical oncology, and several clinically important biomarkers are implemented in routine practice. In an effort to guarantee high quality of molecular testing in France, three subsequent external quality assessment rounds were organized at the initiative of the National Cancer Institute between 2012 and 2014. The schemes included clinically relevant biomarkers for metastatic colorectal (KRAS, NRAS, BRAF, PIK3CA, microsatellite instability) and non-small cell lung cancer (EGFR, KRAS, BRAF, PIK3CA, ERBB2), and they represent the first multigene/multicancer studies throughout Europe. In total, 56 laboratories coordinated by 28 regional molecular centers participated in the schemes. Laboratories received formalin-fixed, paraffin-embedded samples and were asked to use routine methods for molecular testing to predict patient response to targeted therapies. They were encouraged to return results within 14 calendar days after sample receipt. Both genotyping and reporting were evaluated separately. During the three external quality assessment rounds, mean genotype scores were all above the preset standard of 90% for all biomarkers. Participants were mainly challenged in case of rare insertions or deletions. Assessment of the written reports showed substantial progress between the external quality assessment schemes on multiple criteria. Several essential elements such as the clinical interpretation of test results and the reason for testing still require improvement by continued external quality assessment education.

While surgery is the usual treatment for localized well-differentiated and dedifferentiated liposarcomas (WDLPS/DDLPS), the therapeutic options for patients with advanced disease are limited. The classical antimitotic treatments are most often inefficient. The establishment of genetically characterized cell lines is therefore crucial for providing in vitro models for novel targeted therapies. We have used spectral karyotyping, fluorescence in situ hybridization with whole chromosome painting and locus-specific probes, and array-comparative genomic hybridization to identify the chromosomal and molecular alterations of a novel cell line established from a recurring sclerosing WDLPS. The karyotype was hypertriploid and showed multiple structural anomalies. All cells retained the presence of a giant marker chromosome that had been previously identified in the primary cell cultures. This giant chromosome contained high-level amplification of chromosomal regions 12q13-21 and lacked the alpha-satellite centromeric sequences associated with WDLPS/DDLPS. The 12q amplicon was large, containing 370 amplified genes. The DNA copy number ranged from 3 to 57. The highest levels of amplification were observed at 12q14.3 for GNS, WIF1, and HMGA2. We analyzed the mRNA expression status by real-time reverse transcription polymerase chain reaction for six genes from this amplicon: MDM2, HMGA2, CDK4, TSPAN31, WIF1, and YEATS4. mRNA overexpression was correlated with genomic amplification. A second amplicon originating from 10p11-14 was also present in the giant marker chromosome. The 10p amplicon contained 62 genes, including oncogenes such as MLLT10, previously described in chimeric fusion with MLL in leukemias, NEBL, and BMI1.

No abstract

Sequencing of genes encoding mitogen-activated protein kinase (MAPK) pathway proteins in colorectal cancer (CRC) has established as dogma that of the genes in a pathway only a single one is ever mutated. We searched for cases with a mutation in more than one MAPK pathway gene (co-mutations). Tumor tissue samples of all patients presenting with CRC, and referred between 01/01/2008 and 01/06/2015 to three French cancer centers for determination of mutation status of RAS/RAF+/-PIK3CA, were retrospectively screened for co-mutations using Sanger sequencing or next-generation sequencing. We found that of 1791 colorectal patients with mutations in the MAPK pathway, 20 had a co-mutation, 8 of KRAS/NRAS, and some even with a third mutation. More than half of the mutations were in codons 12 and 13. We also found 3 cases with a co-mutation of NRAS/BRAF and 9 with a co-mutation of KRAS/BRAF. In 2 patients with a co-mutation of KRAS/NRAS, the co-mutation existed in the primary as well as in a metastasis, which suggests that co-mutations occur early during carcinogenesis and are maintained when a tumor disseminates. We conclude that co-mutations exist in the MAPK genes but with low frequency and as yet with unknown outcome implications.

No abstract