Background and Rationale for the Childhood Cancer Survivor Study (CCSS) Over the last several decades, advances in treatments for childhood and adolescent cancer have substantially improved survival following diagnosis. These improvements gave rise to the responsibility for investigating long-term treatment-associated morbidity and mortality. Early efforts to describe late effects were largely conducted through single-institution and limited consortia studies. However, by the mid-1980s, it became increasingly clear that these approaches had inherent limitations, including small sample size, convenience sampling, incompletely characterized populations, and limited length of follow-up. To overcome these limitations, the CCSS was proposed and funded by the National Cancer Institute (NCI) as a U01 grant in 1994. Subsequently, the strengths of the CCSS, including an efficient and extensive infrastructure, plus expanding database and biorepository, were recognized and appreciated. Thus, in consultation with the NCI, the CCSS was converted to a U24 (resource grant) funding mechanism to serve the scientific community in 2000. The overarching goal of the CCSS resource is to increase the conduct of innovative and high impact research related to pediatric cancer survivorship. CCSS has been used extensively by researchers from a wide range of disciplines to address a broad spectrum of topics. Strengths of the resource include its large size, comprehensive annotation of treatment exposures, ongoing longitudinal follow-up with characterization of a wide array of participant characteristics and outcomes, and an established biorepository. Design of the Childhood Cancer Survivor Study The Childhood Cancer Survivor Study (CCSS) is a multi-institutional, multi-disciplinary collaborative research resource comprised of a retrospective hospital-based cohort of survivors of childhood cancer and a comparison sibling cohort. Eligible survivors from 31 participating institutions were diagnosed between 1970 and 1999, prior to age 21 years, with selected common pediatric cancers (leukemia, central nervous system tumors, Hodgkin lymphoma, non-Hodgkin lymphoma, kidney tumors, neuroblastoma, soft tissue sarcoma, or bone tumors). All patients who survived five years from the date of diagnosis were eligible, regardless of disease or treatment status. The baseline questionnaire was completed by 24,368 survivors and 5,039 siblings recruited to serve as a comparison group. To date, participants have completed three general follow-up surveys, as well as a number of specialized surveys on specific topics (e.g. health care, insurance, screening practices, men's and women's health issues, adolescent health, sleep and fatigue). In addition, biological samples (buccal cells, saliva and/or blood) have been collected for over 11,000 participants. Full descriptions of the design and characteristics of the CCSS have been previously published (Robison et al; Leisenring et al.), and available data and samples are described at https://ccss.stjude.org/develop-a-study/gwas-data-resource.html. Treatment Data in the Childhood Cancer Survivor Study A key feature of CCSS is the availability of detailed treatment data, which were collected by abstraction of medical records for each individual member of the cohort. Detailed abstraction included dates of therapy, protocol information, and specific details regarding surgery, chemotherapy and radiation. Quantitative dose details were collected for 22 specific chemotherapeutic agents, including alkylating agents, anthracyclines, platinum compounds and epipodophyllotoxins. In addition to individual agent doses, algorithms have been created to calculate cumulative doses of all drugs in a specific class, such as anthracyclines (doxorubicin, daunomycin and idarubicin) or platinum agents (cisplatinum and carboplatinum). Data abstracted for surgeries included dates and both the names and corresponding International Classification of Diseases (9th revision) code. For radiation treatment data, all relevant records were sent to the Radiation Physics Center at M.D. Anderson Cancer Center for detailed abstraction and dosimetry. Initial body region dosimetry was performed for all participants, followed by more detailed dosimetry as needed for specific studies. Genomics Data in the Childhood Cancer Survivor StudyThe NCI's Division of Cancer Epidemiology and Genetics and CCSS investigators collaborated to conduct genomics studies (SNP array genotyping and whole exome sequencing) using samples from the CCSS Biorepository. Studies included all cohort participants with available DNA regardless of sex or ancestry when the genomics studies were initiated. Phenotype Data in the Childhood Cancer Survivor Study Vital status and cause of death for both participants and non-participants is determined via linkage with the National Death Index (NDI). Identification of subsequent neoplasms is based on self-report, followed by validation using medical records, or via NDI. A wide array of additional health outcomes have been ascertained via a comprehensive set of questions on the CCSS questionnaires, covering potential adverse events across a range of organ systems (hearing/vision/speech, urinary, hormonal, heart and circulatory, respiratory, digestive, brain and nervous systems). In addition to health outcomes, longitudinal data have been collected on demographics, health behaviors, family history, screening practices, insurance status, and a range of psychosocial and neurocognitive factors. A full listing of available variables and copies of the CCSS questionnaires are available at http://ccss.stjude.org. Research Areas in the Childhood Cancer Survivor Study Extensive use by the research community has resulted in over 265 published manuscripts on a wide range of topics, including associations between treatment factors and mortality, subsequent neoplasms, chronic health conditions, cardiac events, neurocognitive sequelae, psychosocial factors, fertility, and health status. Additional topics have included health behaviors, screening practices, health care access and utilization, statistical and exposure assessment methodology, and development of risk prediction models. A full listing of published manuscripts using CCSS data is available on the CCSS website at https://ccss.stjude.org/published-research/publications.html. The Childhood Cancer Survivor Study as a Resource for Investigators The CCSS is an NCI-funded resource (U24 CA55727) to promote and facilitate research among long-term survivors of cancer diagnosed during childhood and adolescence. Interested investigators are encouraged to develop research ideas and propose projects within CCSS, whether or not they are from a participating CCSS institution. The CCSS is now accepting proposals to collaborate with CCSS and NCI investigators in the use of genomics data and corresponding outcomes-related data to address innovative research questions relating to potential genetic contributions to risk for treatment-related outcomes. Any researcher, or group of researchers, qualified to conduct genetic research can submit a proposal. There are no restrictions relative to country, institution, or prior involvement in CCSS. A full description of the process for developing a proposal for genetic research in CCSS can be found at https://ccss.stjude.org/develop-a-study/gwas-data-resource.html, along with listings of approved proposals.
Genome-wide association studies (GWAS) of colorectal cancer (CRC) have been instrumental in identifying a number of common susceptibility loci in Non Hispanic (NH)-White populations, and a NCI priority is to extend GWAS findings to other populations to address racial/ethnic disparities in cancer susceptibility. Currently, GWA studies of CRC in NH-Whites, Japanese and African-Americans are ongoing. We propose a complementary study to address this critical research area in Hispanics. Hispanics represent the fastest growing ethnic population in the U.S. and have been largely understudied in terms of genetic susceptibility to cancer. There are noted differences in incidence, survival and mortality in CRC by ethnic/racial groups. Hispanics often present with CRC at a younger age and have a significantly greater incidence of stage IV tumors or metastatic disease compared to NH-Whites. We propose to conduct a large, cost-efficient, population-based GWAS in Hispanics by building upon existing NIH-funded resources, the Colon Cancer Family Registry (Colon CFR) and the Multiethnic Cohort Study (MEC). We plan to recruit 2,500 Hispanic men and women diagnosed with CRC between 01/2008 to present using cancer registries in California, physican referrals and familial referrals. Risk factor/diet questionnaires, pathology reports, Oragene saliva samples (for genotyping), optional blood samples (for genotyping and biometric analysis) and tumor blocks (for MSI testing) will be collected using methodologies developed in the Colon CFR/MEC. Cases of CRC in the MEC (currently 473; anticipated 600 at end) will also be included. Population-based Hispanic individuals without a diagnosis of CRC participating in other GWA studies in the MEC (n=3,900, U01HG004726, Haiman) will be used as controls. We will genotype all 3,100 cases using the Illumina 1M array and use available genotype and epidemiologic data collected on 3,900 controls. Our statistical analyses will include: single-SNP and haplotype effects, gene-environment interactions and heterogeneity by MSI, tumor subtype and family history of CRC. We will replicate findings in a second-stage using CRC cases and controls from Mexico (1,000 cases and 1,000 controls, EU FP7 funding, CHIBCHA, Carvajal-Carmona/Tomlinson). We will also examine heterogeneity of the risk estimates by ethnicity/race by leveraging GWA data on NH-Whites (2,142 cases, 1,909 controls, U01 CA122839, Casey), (4,000 cases, 6,000 NH-White controls, UK-CHIBCHA, Tomlinson), Colombians (2,000 cases and 2,000 controls, CHIBCHA), Japanese (1,000 cases and 1,000 controls) and African-Americans (1,500 cases and 1,500 controls, R01CA126895, Le Marchand). We will genotype replicated significant SNPs in our main and combined analysis in several Hispanic populations (note: studies funded by EU or NIH for data collection but not GWAS), including 800 Puerto Ricans, 2,000 Brazilians, 2,000 Argentineans and 3,000 Spanish/Portuguese, to assess generalizability of findings. We will examine the differences in inflammatory gene transcription dynamics in leukocytes (from blood sample collection) by fatigue level (as assessed from study questionnaire data). This study will have a high impact by addressing the key question of racial/ethnic disparities related to genetic susceptibility to CRC, will provide translational guidelines on biological mechanisms during the cancer survivorship period to increase quality of life among cancer survivors, and will enable further growth and investment into research among Hispanics by providing a resource of genetic data and biospecimens, which is lacking.
To identify mutations in CML
Sequencing of drug resistant organoids
Targeted sequencing of haematopoietic colonies.
Most prostate cancer deaths are caused by metastatic, castration resistant disease (mCRPC). To develop a precision medicine framework for mCRPC, we established a multi-institutional, international clinical sequencing infrastructure to enroll and carry out prospective whole exome and transcriptome sequencing of tumors from a cohort of mCRPC patients. We obtained high quality DNA and RNA sequence data from 150 bone or soft tissue biopsies. Central pathology revealed high-grade adenocarcinoma with only four cases (3.6%) showing neuroendocrine differentiation. Aberrations of AR, ETS genes, TP53 and PTEN were frequent (40-60% of cases), with TP53 and AR alterations being the most enriched in mCRPC compared to primary prostate cancer. We identified novel genomic alterations in PIK3CA/B (fusions and mutations); R-spondin, BRAF and RAF1 (fusions); APC (inactivating mutations); delta-catenin (missense mutations); and ZBTB16/PLZF (homozygous deletions). Aberrations of BRCA2, BRCA1 and ATM were observed at substantially higher frequencies (19.3% overall) than seen in primary prostate cancers, with 56% of these being exclusively somatic. Putative driver gene alterations were identified in nearly all patients, and over half also harbored driver gene fusions, homozygous deletions and/or amplifications. Moreover, 89% of patients harbored a clinically actionable aberration including 62.7% with aberrations in AR, 65% in other cancer-related genes, and 8% with actionable pathogenic germline alterations. Overall, integrative clinical sequencing analysis can be safely and efficiently performed in mCRPC, yields findings that may be actionable for enrolling patients in clinical trials of targeted therapies, and may inform the basis of individual clinical responses.
Somatic copy number alterations (SCNAs) can be detected in tumor cell-free DNA (cfDNA) by shallow whole genome sequencing (sWGS). Polymerase chain reaction (PCR) is typically included in library preparations, but a PCR-free method could be an effective alternative for research and diagnostics. Peripheral blood samples (n=22) were collected in EDTA-containing tubes from non-small cell lung cancer patients (n=10) and healthy donors (n=12). PCR and PCR-free library preparations underwent sWGS and the following metrics were compared: mapping quality, proportion of duplicate reads, genome coverage, sequencing read length and copy number profiles. The PCR-free method was further evaluated on archived plasma samples (n=33) of varying pre-analytical quality collected in EDTA or lithium heparin tubes. Archived samples were from acute myeloid leukaemia (n=15) and B-cell lymphoma (n=18) patients. The percentage of unique reads was significantly higher for the 22 samples analysed by PCR-free (95.7%) compared to PCR (84%). PCR-free libraries from 33 archived samples produced data of a comparable quality, with 95.4% unique reads. All other evaluated metrics were highly comparable for PCR and PCR-free library preparations. We conclude that SCNA detection can be performed by PCR-free sWGS on cfDNA from cancer patients and is feasible for diagnostic and archival plasma collected in EDTA- or lithium heparin-containing tubes. Our results pave the way for an automated cfDNA workflow for cancer plasma samples.
