Author of

• 11344

  +

  MO10 - Molecular Pathology II (ID 127)

  • Event: WCLC 2013
  • Type: Mini Oral Abstract Session
  • Track: Pathology
  • Presentations: 1
  • Moderators:W.A. Franklin, A. Mahar
  • Coordinates: 10/28/2013, 16:15 - 17:45, Bayside 104, Level 1

  • 11356

    +

    MO10.12 - ROS1 Fusions Diagnosed by Break-Apart FISH in NSCLC (ID 968)

    17:20 - 17:25  |  Author(s): A. Le
    • Abstract
    • Presentation
    • Slides

    Background
    Chromosomal rearrangements which generate constitutively activated ROS1 receptor tyrosine kinase (6q22.1) have been found in several tumor types, including non-small cell lung cancers (NSCLC). In clinical trials, the oral kinase inhibitor crizotinib has shown promise in treating tumors with ROS1 rearrangements. Currently, fluorescence in situ hybridization (FISH) using dual-color, break-apart (BA) probes is used to detect ROS1 rearrangements in clinical samples; however, further optimization of this method is necessary to ensure patients are accurately diagnosed. This study explores BA FISH assay characteristics in NSCLC samples.

    Methods
    Tumor sections from 464 NSCLC patients were screened for ROS1 rearrangement using ROS1 BA FISH. Of these samples, 206 were co-screened for ALK rearrangement. The copy number of fused and isolated 3’/5’ signals, as well as the incidence of atypical patterns (doublet and clustered multiple fusions) was investigated. Cells were considered ROS1 positive (ROS1+) when ≥ 15% of nuclei displayed split 5’/3’ signals or single 3’ signals. Specific fusion transcripts in ROS1+ cases were identified by RT-PCR or inverse PCR.

    Results
    ROS1 rearrangements (ROS1+) were found in 21 patients (5%). The copy number of native ROS1 differed significantly between positive and negative tumors (mean of 1.5 versus 2.5, p<0.0001). The percent of cells with FISH patterns compatible with ROS1 rearrangement ranged from 30% to 100%, with a mean of 81%, in ROS1+ patients. The distribution of positive cells between scored regions within ROS1+ tumors was investigated for 13 cases and found to follow a normal distribution, ruling out intra-tumoral heterogeneity. Among ROS1+ specimens, 71% had a split signal pattern, 19% displayed a single 3’ pattern, and 10% had both a split and single 3’ pattern of positivity. For positive tumors, ROS1 fusion partners were identified as SDC4 (S2;R32 and S2;R34), EZR (E10;R34) and CD74 (C6;R32 and C6;R34). Atypical negative patterns such as fused doublets, clusters, 3’ doublets, 5’ doublets, and single 5’ signals were observed in 4%, 1%, 1%, <1%, and <1% of negative patients. ALK and ROS1 were scored simultaneously in the same cells in 206 patients, including 5 ROS1+ and 10 ALK+; no double positive cases were found. In ROS1 negative specimens, mean native ALK copy numbers were significantly higher than native ROS1 in ALK negative samples (3.2 versus 2.3, p<0.0001).

    Conclusion
    ROS1+ tumors were detected in 5% of patients in this large NSCLC cohort. Since these patients were subject to various selection strategies, this frequency cannot be transferred to an unselected NSCLC population. The low native ROS1 copy number in the rearranged cells and lack of evidence of intra-tumoral heterogeneity suggests ROS1 rearrangements occur early in tumorigenesis, consistent with their known oncogenic driver role. Data from this sample also show that, in FISH negative cases, ROS1 copy number was lower than native ALK. This suggests ROS1 may exist in a relatively more stable portion of the genome, potentially explaining why ROS1 rearrangements exist at a lower frequency than ALK rearrangements in NSCLC.

    Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.