Home >> ALL ISSUES >> 2018 Issues >> Molecular Pathology Abstracts, 2/18

Molecular Pathology Abstracts, 2/18

Print Friendly, PDF & Email

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.

Gene expression and risk of leukemic transformation in myelodysplasia

The myelodysplastic syndromes represent a group of clonal hematopoietic disorders with varying prognoses, with survival ranging from a few months to more than 10 years. Multiple laboratory measurements have been used in attempts to provide reliable prognostic assessments, including bone marrow blast counts, severity of peripheral cytopenia, cytogenetic findings, and, most recently, gene-mutation profiling. However, no method has been found to be completely accurate. The clinical use of gene-expression profiling for prognostic evaluation is becoming increasingly common for multiple types of solid tissue tumors but has not found significant application in cases of myelodysplastic syndrome (MDS). A recent study by the authors provides evidence that gene-expression profiling of MDS, alone or in combination with current modalities, may significantly improve the prognostic evaluation of these conditions. The authors first used transcriptome sequencing to evaluate gene-expression patterns in bone marrow CD34+ cells from 100 patients with MDS, none of whom had received treatment other than supportive care prior to the analysis. They identified 3,142 genes for which expression levels varied significantly in the 100 samples. Clustering analysis revealed two subgroups based on differential expression levels, with further subdivision found to be nonstable. The subgroups displayed distinct hematological and transcriptomic profiles, with the first subgroup showing increased expression of genes specific to megakaryocyte/erythrocyte progenitors (designated the EMK group) and the second group showing increased expression of genes associated with immature hematopoietic progenitors (the IMP group). A comparison of survival rates demonstrated that the IMP group was significantly associated with inferior overall survival compared with the EMK group. In addition, all patients who ultimately progressed to leukemic transformation were in the IMP group. To facilitate the clinical use of these findings, the authors next used regression analysis of the expression data to identify a 68-gene subgroup of the 3,142 genes that could be used to predict the two classifier groups using a microarray assay. The microarray assay was used on an external cohort of 183 MDS patient samples to classify each as belonging to the EMK or IMP subgroup. In agreement with the previous findings, patients from this cohort who were assigned to the IMP subgroup had significantly shorter survival rates than those in the EMK subgroup. Finally, the authors assessed whether incorporating their expression profile data would improve the performance of the Revised International Prognostic Scoring System (IPSS-R) for prognostic prediction. An analysis of samples from 148 patients with MDS and complete clinical data found that adding expression data to the IPSS-R categories significantly improved the identification of patients with a high risk of leukemic transformation among those categorized as low risk by the IPSS-R algorithm. Although this study is not the first of its kind, it does demonstrate the feasibility of using gene-expression profiling in the clinical evaluation of MDS and supports the potential value of this information in the prognostic assessment of MDS.

Shiozawa Y, Malcovati L, Galli A, et al. Gene expression and risk of leukemic transformation in myelodysplasia. Blood. 2017;130(24): 2642–2653.

Correspondence: Mario Cazzola at mario.cazzola@unipv.it and Dr. Seishi Ogawa at sogawa-tky@umin.ac.jp


Co-occurring genetic alterations in advanced-stage EGFR-mutant lung cancers

The current dogma of cancer genetics holds that tumorigenesis occurs via acquisition of a somatic mutation in an oncogene and alters the activity of the encoded protein in such a way that promotes abnormal growth, differentiation, or survival. This process is typically the result of a functionally significant driver mutation developing in a single gene, which is then viewed as a potential target for therapeutics. The authors suggest that this model is likely an oversimplification of reality, providing evidence that the co-occurrence of multiple, functionally significant genetic lesions is more common than has been recognized in the past. The authors analyzed cell-free DNA (cfDNA) samples from a group of 1,122 EGFR-mutation-positive and 944 EGFR-mutation-negative patients with stage III or IV nonsmall cell lung cancer (NSCLC) using a next-generation sequencing assay covering 68 clinically relevant cancer genes. Most (92.9 percent) of the EGFR-mutation-positive patients had at least one additional variant of known or likely functional significance, in addition to the EGFR driver mutation. Of the additional variants observed in the mutation-positive group, 89.8 percent (3,033 of 3,375 mutations observed) had verified or likely functional effects.The remaining 10.2 percent were classified as likely passenger mutations. Of the mutations present in the EGFR-mutation-negative group, 16.1 percent (415 of 2,578 mutations observed) were classified as likely passenger events. EGFR driver mutations were found to co-occur with additional driver mutations in several other genes, including PIK3CA, BRAF, MET, MYC, CDK6, and CTNNB1. Alterations in these genes were also significantly enriched in the EGFR-mutation-positive group relative to the EGFR-mutation-negative group. An analysis of a subgroup of EGFR-mutation-positive cases that contained a resistance conferring p.Thr790Met mutation (n = 440) or p.Cys797Ser mutation (n = 15) demonstrated that these resistance-mutation-positive samples were enriched for co-occurring alterations in the MAPK-pathway genes, cell cycle pathway genes, and genes encoding hormone signaling proteins, suggesting a functional, cooperative role in driving resistance to EGFR-targeted therapy. In addition, an analysis of samples from a cohort of patients treated with a tyrosine kinase inhibitor demonstrated that those responding to treatment (n = 37) had significantly fewer co-occurring mutations (mean, 2.7) than those who did not respond (n = 36; mean, 5.5 mutations). Finally, a single NSCLC case was followed longitudinally over six years, spanning initial diagnosis, metastatic progression, tyrosine kinase inhibitor therapy, and eventual evolution and patient death using whole exome sequencing of tumor tissue as well as cfDNA analysis. Results from the whole exome sequencing analyses of this case recapitulated the findings from the broader study using cfDNA samples in that co-occurring mutations were found in the same mechanistic pathways. This study’s finding of multiple co-occurring genetic alterations in the majority of advanced stage EGFR-mutation-positive lung cancers serves to illuminate the complexity of these tumors and suggests that a new model of oncogenesis that accounts for this complexity may be necessary to derive maximal therapeutic benefit.

Blakely CM, Watkins TB, Wu W, et al. Evolu­tion and clinical impact of co-occurring genetic alterations in advanced-stage EGFR-mutant lung cancers. Nat Genet. 2017;49(12):1693–1704.

Correspondence: Dr. Trever G. Bivona at trever.bivona@ucsf.edu or Dr. Charles Swanton at charles.swanton@crick.ac.uk


Check Also

Clinical pathology selected abstracts

February 2019—Impact of commercial laboratory testing on a care delivery system: Care delivery systems have become increasingly fragmented and complex, which impacts patient care. The amount of health care data generated has also created a problem by reducing the time devoted to direct patient care.