Abstract 5607: Copy-number profiling of single, digitally-sorted PD-L1 positive cells in NSCLC FFPE tissues

Abstract

Abstract Introduction: Immune system strongly affects cancer cell fate; indeed, immune-checkpoint blockade is emerging as an effective therapeutic strategy in a broad range of tumor types. The effectiveness of these therapies depends on different aspects of tumor biology such as the expression of immune checkpoint proteins, the presence of infiltrating active immune cells and many others. To date, a robust biomarker to predict therapeutic response to these drugs has not been found yet. Among all the strategies used to identify effective biomarkers, a single-cell level molecular profiling of tumor, stromal and infiltrating cells, has not been explored due to technical challenges. Here we present a method for the digital isolation of single PD-L1+ cells from FFPE tissues, along with their genetic characterization for the identification of novel biomarkers. Methods: FFPE tissue sections from two NSCLC patients, were independently tested for PD-L1 expression using IHC VENTANA PD-L1 (SP142) Assay and scored in terms of tumor content (30%, 85%), percentage of PD-L1 positive tumor cells (70%, 95%), PD-L1 intensity, and H-score (130, 175). From the same tissue blocks, additional FFPE sections were dissociated down to cell suspension, and immunofluorescently labelled using Keratin-AlexaFluor®488, Vimentin-AlexaFluor®647 and PD-L1 (clone SP263)-AlexaFluor®546. Four cell populations (Keratin+PD-L1-, Keratin+PD-L1+, Vimentin+PD-L1- and Vimentin+PD-L1+) were identified at the DEPArray™ platform and, for each, we isolated pools of precise number (range 43-121) of cells, along with pure single cells. Recovered cells were whole genome amplified (Ampli1™ WGA), and genome-wide copy-number aberrations (CNA) profiles were obtained by Ampli1™ LowPass kit and IonTorrent sequencing. Results: The expression of PD-L1 on cell membrane was attributed -by means of co-expression of either keratin or vimentin- to tumor or stromal cells, respectively. For both patients, in the tumor populations numerous gains and losses were clearly identified along the genome; Keratin+PD-L1+ and Keratin+PD-L1- pooled cells displayed comparable profiles. Notably, in both patients copy gains were identified on chromosome 9 region including the PD-L1 gene. A variable range of alleles (3 to 6) on the same PD-L1 locus was detected on single tumor cells, along with a high inter-cell heterogeneity in terms of gain extent and genomic regions involved, suggestive of the presence of different tumor clones, hidden in pooled cells data. All the stromal (Vimentin+PD-L1+/-) cells showed flat profiles. Conclusions: This study demonstrates an innovative approach for isolation and genetic profiling of single pure PD-L1+ cells from FFPE tissue. DEPArray™ sorting combined with low-pass whole genome sequencing enables high quality genome-wide profiling of pure single cells and pools isolated from phenotypically distinct populations in FFPE tumor tissue. Citation Format: Giulio Bassi, Chiara Bolognesi, Chiara Mangano, Claudio Forcato, Alberto Ferrarini, Valentina Del Monaco, Paola Tononi, Genny Buson, Gianni Medoro, Manaresi Nicolò, Francesca Fontana. Copy-number profiling of single, digitally-sorted PD-L1 positive cells in NSCLC FFPE tissues [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5607. doi:10.1158/1538-7445.AM2017-5607