Paul M. Williams

Robustness of Next Generation Sequencing on Older Formalin-Fixed Paraffin-Embedded Tissue

Next Generation Sequencing (NGS) technologies are used to detect somatic mutations in tumors and study germ line variation. Most NGS studies use DNA isolated from whole blood or fresh frozen tissue. However, formalin-fixed paraffin-embedded (FFPE) tissues are one of the most widely available clinical specimens. Their potential utility as a source of DNA for NGS would greatly enhance population-based cancer studies. While preliminary studies suggest FFPE tissue may be used for NGS, the feasibility of using archived FFPE specimens in population based studies and the effect of storage time on these specimens needs to be determined. We conducted a study to determine whether DNA in archived FFPE high-grade ovarian serous adenocarcinomas from Surveillance, Epidemiology and End Results (SEER) registries Residual Tissue Repositories (RTR) was present in sufficient quantity and quality for NGS assays. Fifty-nine FFPE tissues, stored from 3 to 32 years, were obtained from three SEER RTR sites. DNA was extracted, quantified, quality assessed, and subjected to whole exome sequencing (WES). Following DNA extraction, 58 of 59 specimens (98%) yielded DNA and moved on to the library generation step followed by WES. Specimens stored for longer periods of time had significantly lower coverage of the target region (6% lower per 10 years, 95% CI: 3-10%) and lower average read depth (40x lower per 10 years, 95% CI: 18-60), although sufficient quality and quantity of WES data was obtained for data mining. Overall, 90% (53/59) of specimens provided usable NGS data regardless of storage time. This feasibility study demonstrates FFPE specimens acquired from SEER registries after varying lengths of storage time and under varying storage conditions are a promising source of DNA for NGS.

Abstract CT101: NCI-molecular analysis for therapy choice (NCI-MATCH) clinical trial: interim analysis

The NCI MATCH is an efficacy signal-finding trial to identify molecular alterations that can be matched to targeted treatments in tumor biopsies from patients (pts) with refractory solid tumors or lymphomas.br />METHODS: An Ion Torrent Oncomine panel sequencing assay and a binary PTEN immunohistochemistry assay were employed. NCI-MATCH opened in August 2015. A protocol-directed pause in screening for interim analysis occurred in November 2015 with goals to assess tumor biopsy feasibility and quality, performance of the CLIA-accredited laboratories, any unanticipated concerns with the study, the match rate for the first activated 10 arms and for planned arms, the spectrum of molecular abnormalities identified, histologic tumor types and demographics. RESULTS: Between 8-12-15 and 11-11-15, 795 pts enrolled for screening and 739 biopsies were submitted. Biopsies were submitted from community (2/3) and academic (1/3) sites. Sequencing was completed on 645 of these specimens (87%). Median turn around time was 27 days but increased from 14 days in the first month to 36 days in the last month, correlating with marked increase in weekly accrual. The highest toxicity (Grade 3) possibly related to biopsy was < 1%. The most frequent tumor types sequenced were colorectal (13%), breast (13%), ovarian (11%), non-small cell lung (7.4%), endometrial (7%), pancreatic (5%), head/neck (5%) esophageal/gastric (4%), cholangiocarcinoma and neuroendocrine (3% each), and small cell lung, prostate, bladder, and unknown primary (<3% each). Fifty-six patients (9%) had actionable mutations of interest (aMOIs); of these, 33 pts (5.1%, 95%CI 3.5, 7.1%) met molecular eligibility for assignment to one of 10 treatment arms. The most common genetic mutations, amplifications or translocations were: CONCLUSIONS: Accrual was brisk and biopsies were feasible and safe with adequate tumor yield and results. The match rate was lower than predicted for the first 10 treatment arms. Accrual of less common tumor histologic types exceeded expectations. Results from the interim analysis will inform the types of treatment and predicted match rate for additional MATCH arms. NCI-MATCH will re-open with increased capacity and additional treatment arms with a match rate goal of at least 20%. Citation Format: Barbara A. Conley, Robert Gray, Alice Chen, Peter O’Dwyer, Carlos Arteaga, Brent Coffey, David Patton, Shuli Li, Lisa M. McShane, Larry Rubinstein, Robert Comis, Jeffrey Abrams, Paul M. Williams, Chih-Jian Lih, Stanley Hamilton, Edith Mitchell, James Zwiebel, Keith Flaherty, NCI MATCH team. NCI-molecular analysis for therapy choice (NCI-MATCH) clinical trial: interim analysis. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr CT101.

Criteria for use of omics-based predictors in NCI-sponsored clinical trials.

58 Background: High-throughput omics technologies (e.g., genomics, epigenomics, proteomics, metabolomics) offer exciting opportunities for new biological insights into cancer. The IOM report on translational omics defined omics as the study of related sets of biological molecules in a comprehensive fashion. (IOM (Institute of Medicine) 2012. Evolution of Translational Omics: Lessons Learned and the Path Forward. Washington, DC: The National Academic Press.) The promise of omics technologies has proven problematic to translate into clinically useful tests. Difficulty obtaining biospecimens, unrecognized preanalytical influences, and suboptimal assay analytical performance can lead to unreliable results and conflicting reports. Poor reporting of study details and limited access to data and computer code can thwart efforts to replicate published results or to detect flaws in study design and analysis methods. METHODS NCI held an interactive workshop for a wide variety of stakeholders to explore better approaches to omics-based test development and validation. This workshop heavily informed the ideas presented here. Recommendations are stated concisely, then explained. RESULTS A checklist of items to consider when evaluating the evidence for clinical use of an omics-based predictor, including in a trial where it will guide therapy, is presented. It covers specimen and assay requirements, the predictor model development process, clinical study design and conduct, and regulatory, ethical, and legal issues. The list applies to any trial involving investigational use of an omics test that will alter the clinical management of patients. The criteria also largely apply to situations in which the test will be evaluated retrospectively on specimens collected from patients who were prospectively enrolled on clinical studies. CONCLUSIONS The proposed checklist should serve as a useful guide to investigators planning to submit proposals for NCI-funded studies involving use of an omics-based test. Ideally, this checklist will be consulted in the assay planning and development phases so that the necessary evidence will have been collected in a well-documented fashion by the time definitive evaluation of the test is desired.

Expression array analysis of the hepatocyte growth factor invasive program

Abstract Signaling by human hepatocyte growth factor (hHGF) via its cell surface receptor (MET) drives mitogenesis, motogenesis and morphogenesis in a wide spectrum of target cell types and embryologic, developmental and homeostatic contexts. Oncogenic pathway activation also contributes to tumorigenesis and cancer progression, including tumor angiogenesis and metastasis, in several prevalent malignancies. The HGF gene encodes full-length hHGF and two truncated isoforms known as NK1 and NK2. NK1 induces all three HGF activities at modestly reduced potency, whereas NK2 stimulates only motogenesis and enhances HGF-driven tumor metastasis in transgenic mice. Prior studies have shown that mouse HGF (mHGF) also binds with high affinity to human MET. Here we show that, like NK2, mHGF stimulates cell motility, invasion and spontaneous metastasis of PC3M human prostate adenocarcinoma cells in mice through human MET. To identify target genes and signaling pathways associated with motogenic and metastatic HGF signaling, i.e., the HGF invasive program, gene expression profiling was performed using PC3M cells treated with hHGF, NK2 or mHGF. Results obtained using Ingenuity Pathway Analysis software showed significant overlap with networks and pathways involved in cell movement and metastasis. Interrogating The Cancer Genome Atlas project also identified a subset of 23 gene expression changes in PC3M with a strong tendency for co-occurrence in prostate cancer patients that were associated with significantly decreased disease-free survival.

Abstract 1039: PDX models generated from a patient with metastatic colon adenocarcinoma display both spatial and temporal tumor heterogeneity

Background: Patient-derived Xenograft (PDX) models are being widely used in preclinical studies to identify biomarkers of drug response and to enhance our understanding of cancer biology. Since patients with metastatic cancer have both intra-tumor and inter-site heterogeneity, PDX models generated from different tumor sites may provide a way to study tumor heterogeneity. Characterization of the genomic landscape in these models may also provide better insights into treatment response or resistance. It is rare to have multiple PDX models generated from a single patient over multiple time points during a treatment trajectory. Here, we report the genomic profiles of PDX models generated from 4 distinct tissue specimens over a 7-month period from a patient with metastatic colon adenocarcinoma. The first 2 PDX models were generated from circulating tumor cells (CTCs) and a liver biopsy prior to treatment with a combination pan-AKT + MEK inhibitor regimen. A third PDX model was generated from a liver biopsy while on-treatment and a fourth from an adrenal gland resection at progression. Clinically, all reported metastatic sites, except the adrenal gland, responded to the combination therapy. Results: Genomic characterization of the specimens obtained from these 4 PDX models led to the following observations: 1) PIK3CA E545K and KRAS G12D are present in all the specimens tested for all 4 models and are likely truncal driver mutations; 2) exclusive inter-model SNVs (single nucleotide variants) were identified, and may be model-specific variants representing inter-site heterogeneity in the patient; 3) variants involved in known resistance mechanisms to MEK inhibition were not present in any specimens; 4) overexpression of AKT3 has been reported as a resistance mechanism to a pan-AKT inhibitor and was observed in the adrenal tissue from the patient but not in any other PDX model derived from this patient; 5) intra-model and inter-model heterogeneity in whole genome CNV (copy number variant) profiles was observed between individual PDXs obtained from the pre-treatment CTC-derived model and the on-treatment liver biopsy model. Interestingly, one of the PDXs from the CTC-derived model presented a sub-clonal tumor fraction closely related to the on-treatment liver biopsy model. The multiple inter-model CNV profiles in the liver biopsy derived PDX models represent temporal heterogeneity within a tissue. Conclusions: We observed genomic heterogeneity in PDXs generated from specimens from a patient with metastatic colon adenocarcinoma. Both truncal and sub-clonal variants were identified representing various tumor fractions in these models. This case study illustrates how genomic profiling of multiple tumor sites at different times during course of treatment can provide insight into the complexity of tumor heterogeneity and tumor evolution in patients with metastatic disease. Citation Format: Biswajit Das, Chris Karlovich, Corrine E. Camalier, Rajesh Patidar, Li Chen, Vivekananda Datta, William D. Walsh, Sean P. McDermott, Tomas Vilimas, Palmer Fliss, Justine N. McCutcheon, Amanda Peach, Michelle Ahalt-Gottholm, Carrie Bonomi, Kelly Dougherty, John Carter, Shivaani Kummar, Yvonne A. Evrard, Melinda G. Hollingshead, Paul M. Williams, James H. Doroshow. PDX models generated from a patient with metastatic colon adenocarcinoma display both spatial and temporal tumor heterogeneity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 1039.

Abstract A053: Prevalence of mismatch repair deficiency (dMMR) in the NCI Molecular Analysis for Therapy Choice (NCI-MATCH or EAY131) population

Introduction: NCI-MATCH (EAY131), a national signal-finding trial, assigns targeted treatments to patients (pts) with relapsed/refractory solid tumors, lymphomas and myelomas based on molecular abnormalities in tumors. We screened 4901 pts for dMMR, an eligibility criterion. Methods: Immunohistochemical expression (IHC) of MLH1 and MSH2 was performed at the CLIA-accredited MD Anderson Clinical IHC Laboratory. Loss of nuclear expression of MLH1 or MSH2 predicted microsatellite instability resulting from sporadic alterations or germline mutations. Assays were validated for FFPE tissue and used under an FDA abbreviated investigational device exemption. IHC was performed on 4 μm deparaffinized and rehydrated FFPE tissue sections. Antigen retrieval was performed at 100°C for 20 minutes with Tris-EDTA buffer solution, pH 6.0 & blocking of endogenous peroxidase with 3% peroxide for 5 minutes. Primary MLH1 antibody (Cell Marque TM , clone G168-728) at 1:300 dilution and primary MSH2 antibody (Calbiochem®, clone FE11) at 1:100 dilution were applied. Primary antibody detection used a commercial polymer system (Bond Polymer Refine Detection, Leica) with stain development using incubation with DAB and DAB Enhancer (Leica). Loss of nuclear expression (required for dMMR eligibility): Complete loss of nuclear expression by tumor cells and retention of staining in non-neoplastic cells (internal control). Intact nuclear expression: Nuclear expression of any intensity within tumor cells. Cannot be determined: Insufficient specimen, technically inadequate IHC assay, or cytoplasmic staining without definite nuclear staining. Results: 4901 pts had MLH1/MSH2 IHC assay, of whom 37 had both MLH1 and MSH2 indeterminate. Of 4864 cases with valid MLH1/MSH2 results, 2% had dMMR (Table–histologies combined to fit allowed space). The most frequent dMMR was in the few thyroid/parathyroid tumors, followed by various gynecologic tumors. Conclusion: 2% of 4864 cases with valid MLH1/MSH2 results in NCI-MATCH had dMMR by IHC; prevalence varied across histologies. Many tumors lacking expression of MLH1 or MSH2 are not prominently associated with Lynch syndrome or frequent somatic dMMR. The NCI-MATCH screen for dMMR adds to information on the potential benefit of screening advanced/refractory tumor types for this abnormality as a standard practice for determining eligibility for PD-1 inhibitors. Citation Format: Barbara A. Conley, Stanley R. Hamilton, Shuli Li, Robert J. Gray, David R. Patton, Peter J. O9Dwyer, Robert L. Comis, Jeffrey S. Abrams, Nilofer S. Azad, Michael J. Overman, Jonathan D. Schoenfeld, Paul M. Williams, James V. Tricoli, Elad Sharon, Howard Streicher, Lyndsay N. Harris, Alice P. Chen, Keith T. Flaherty. Prevalence of mismatch repair deficiency (dMMR) in the NCI Molecular Analysis for Therapy Choice (NCI-MATCH or EAY131) population [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A053.

Abstract 612: The NCI exceptional responders initiatives: Initial feasibility result

Proceedings: AACR 106th Annual Meeting 2015; April 18-22, 2015; Philadelphia, PA Introduction: Preliminary data show that up to 10% of patients on early clinical trials have complete (CR) or partial (PR) responses. Exceptional responses (ER) are also observed in certain clinical cases using standard chemotherapy. The ERI Pilot study is evaluating the molecular alterations found in tumors from cancer patients who respond to a systemic treatment (standard or investigational) that is ineffective in at least 90% of patients. The study is available through the Cancer Trials Support Unit ([www.ctsu.org][1]). Experimental Procedures Cases of patients who achieve a CR or PR > 6 months are proposed by email at NCIExceptionalResponders@mail.nih.gov. Once provisionally approved, the proposer sends tissue to a central repository, where the tissue is evaluated for percent tumor n nucleic acids are isolated & sent for whole exome, mRNA & deep targeted sequencing. Tissue may be frozen or formalin fixed paraffin embedded but must be at least an entire core biopsy. Data will be correlated with mechanism of action of the treatment given. The ERI pilot study will characterize > 100 ERs. If sufficient tissue is available, whole genome sequencing o thus many more tumors will need to be submitted to obtain these results. Six months after the genomic data are generated they will be released to a publicly available, controlled access database to facilitate research by qualified investigators. The ERI study accepts all types of malignant disease in adults or children if adequate tissue obtained prior to the ER exists. The ER could have been to an approved or investigational, targeted, or non-targeted systemic treatment. The feasibility of retrospectively collecting tumor tissue & clinical data for genomic analysis is a goal of the ERI study. Results: The ER study was activated on September 24, 2014. As of December 2, > 80 cases have been proposed, 46 are provisionally approved & 26 cases were not ERs or were withdrawn. Accepted Cases: solid & hematologic tumors (2 brain, 4 breast, 2 colon, 6 gastric, 1 hepatocellular, 7 head/neck, 2 renal, 2 lung cancer, 1 melanoma, 2 ovarian, 6 pancreatic, 2 prostate and 4 sarcomas; 2 acute leukemia, 2 lymphoma, 1 myeloma). Approximately 75% involved standard & 25% involved targeted drugs or combinations. Tissue analysis is ongoing & results will be presented. Conclusions: Collection of archived tissue from ERs for genomic evaluation is feasible. Efficient ER case identification occurs through expert review and consensus-based case selection. Citation Format: Barbara A. Conley, S Percy Ivy, James V. Tricoli, Jean-Claude Zenklusen, Roy Tarnuzzer, Irina Lubensky, Naoko Takebe, Paul M. Williams, JoAnne Zujewski, Richard Little, Jeffrey White, Elise Kohn, Shakun Malik, Ben Kim, Erin Souhan, Lou Staudt. The NCI exceptional responders initiatives: Initial feasibility result. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 612. doi:10.1158/1538-7445.AM2015-612 [1]: http://www.ctsu.org

Moving next-generation sequencing into the clinical realm: Detection of somatic mutations in cancer by targeted amplicon sequencing.

60 Background: Molecular targeted therapies are increasingly important in treating cancer patients; robust analytically validated clinical assays are required for patient selection in early-stage clinical trials. The goal of Molecular Characterization Laboratory (MoCha) is to develop clinical diagnostic assays using next generation sequencing methods to support clinical studies in DCTD (CTEP). METHODS We developed a custom assay for somatic mutation detection using Fluidigm access array technology for amplicon generation followed by sequencing with the Illumina Miseq. A panel of 62 amplicons covering 6 Kb genomic regions was designed to detect 92 DNA loci, including common therapeutically actionable targets, in 37 genes. Analytical studies were performed using genomic DNA samples from fresh or formalin fixed cancer cell-lines and a normal hapmap individual (CEPH). We subsequently applied this assay to characterize DNA samples from both tumor tissues and blood specimens from ovarian cancer patients. RESULTS The assay detected known variants in both frozen and fixed DNA samples reproducibly with high sensitivity and specificity (<2%). Using a series of positive control plasmid spikes mixed into a normal reference CEPH DNA at pre-defined copy number ratios, we verified the assay is sensitive to detect variants at 5% allelic frequency with a minimum 400 X coverage. We identified somatic mutations in TP53 and PIK3CA in a few patients, and a germ-line variant D1583N in ATM genes occurring in one-third of tested patients. CONCLUSIONS We developed and validated a next generation sequencing assay suitable for patient selection for clinical trials. Plans are to correlate sequencing and clinical results when clinical data are available.