Virtual Library

Start Your Search

K. Delisle



Author of

  • +

    MINI 22 - New Technology (ID 134)

    • Event: WCLC 2015
    • Type: Mini Oral
    • Track: Biology, Pathology, and Molecular Testing
    • Presentations: 1
    • +

      MINI22.10 - A New Approach to Large Scale Proteomic Profiling to Uncover Tumor Phenotypes (ID 2166)

      17:40 - 17:45  |  Author(s): K. Delisle

      • Abstract
      • Presentation
      • Slides

      Background:
      Genomic profiling is a powerful method for identifying mutations that drive tumors and matching patients to targeted therapies. However, this may only be a transient solution and resistance commonly emerges as the mechanism of targeted inhibition is overcome. Proteomic profiling of the tumor provides a dynamic tool to survey altered protein expression and deregulated pathways, which in turn may implicate specific treatments or identify novel therapeutic targets. Mass spectrometry offers highly multiplexed proteomic measurements, but extensive sample pre-processing and low sample throughput can lead to extended analysis times of weeks or months. Thus a need exists for a high throughput, sensitive and quantitative platform for proteomic analysis.

      Methods:
      We used the SOMAscan proteomic platform, which measures 1129 proteins with a median limit of detection of 40 fM and 5% CV, to analyze protein lysates from 63 lung tumor samples. The assay does not require sample pre-fractionation, and this study (which generated over 142,000 protein measurements) represents less than one day of SOMAscan throughput. The study consisted of matched tumor/non-tumor protein lysates prepared from 18 squamous cell carcinoma and 45 adenocarcinoma fresh-frozen resected specimens, 86% of which were Stage I/II. The paired log~2~ tumor/non-tumor ratio was calculated and hierarchical clustering heat maps and dendrograms were constructed to identify related protein regions and tumor phenotypes.

      Results:
      Common proteomic changes and unique tumor phenotypic groups were identified by unbiased clustering algorithms. Large, consistent tumor/non-tumor differences of at least 4-fold were observed for 35 proteins in at least 20 (32%) of the tumors. Some of these proteins were more than 100-fold higher in individual tumors. The two most commonly elevated proteins were thrombospondin 2 and MMP12, which were increased in 81% and 61% of the tumors, respectively. We have previously reported higher levels of MMP12 in the serum of lung cancer patients, and the current data supports a tumor-associated origin for circulated MMP12. A second analysis identified sub-phenotypes of tumors clustered by common protein alterations independent of histological classification or mutation status. Many of these tumor subsets had increased expression of known oncology drug targets.

      Conclusion:
      Broad, unbiased high-throughput proteomic profiling of tumor tissue may reveal individual phenotypes that hold the potential to respond to targeted therapies and to monitor therapeutic efficacy throughout treatment. Measuring proteins complements mutation analysis by enabling therapeutic selection beyond driver mutation targets, including immune modulator therapies, repurposing existing drugs and enriching clinical trials with likely responders. While genomics is a fixed snapshot, blood- and tissue-based serial proteomic measurements respond to change and can lead to the personalized adaptation of treatment and identification of novel therapeutic targets.

      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.