SOW_NIHES2015037.pdf

STATEMENT OF WORK

FOR

THE STATISTICAL SUPPORT FOR THE

NATIONAL INSTITUTE OF ENVIRONMENTAL HEALTH SCIENCES

(NIEHS) IN THE DIVISION OF INTAMURAL RESEARCH (DIR)

AND THE NATIONAL TOXICOLOGY PROGRAM (NTP)

Dated December 2, 2015

The National Institute of Environmental Health Sciences (NIEHS) (http://www.niehs.nih.gov/index.cfm), located in Research Triangle Park (RTP), North Carolina, is one of 27 research institutes and centers that comprise the National Institutes of Health (NIH) (http://www.nih.gov/). The mission of the NIEHS is to reduce the burden of human illness and disability by understanding how the environment influences the development and progression of human disease. In-house research at NIEHS is conducted by investigators in the Division of Intramural Research (DIR) and the Division of the National Toxicology Program (DNTP). DIR researchers investigate the role of environmental agents in human disease and dysfunction, underlying mechanisms of environmentally-associated diseases, and intervention/prevention strategies to reduce the effects of exposures to environmental agents. Specific DIR research topics are available at http://www.niehs.nih.gov/research/atniehs/dir/index.cfm. The National Toxicology Program (NTP) (http://ntp.niehs.nih.gov and http://www.niehs.nih.gov/research/atniehs/dntp/index.cfm and) is a federal, interagency program, headquartered at the NIEHS, whose goal is to provide information about potentially hazardous agents by taking advantage of technological advances in molecular biology and computer science.

A. General Requirements

1) Independently, and not as an agent of the government, the contractor shall provide all the necessary services, qualified personnel, material, equipment, and facilities not otherwise provided by the Government as needed to perform the statistical, mathematical, and computational activities described herein.

2) When carrying out the statistical analyses noted below, in consultation with the COR, the contractor shall use a variety of statistical and mathematical methods, which include, but are not limited to, dose-response assessments including estimation of ED50s and benchmark doses, multi-factor analysis of variance, general linear model techniques, multiple comparison procedures, categorical data analyses, mathematical modeling of toxicokinetics and other biological processes, cluster analyses, survival-adjusted analysis of carcinogenicity studies, techniques for normalization and analysis of microarray data from various platforms, non-parametric techniques, and Bayesian methods.

3) As directed by the COR, the contractor shall consult with DNTP and/or DIR staff regarding various statistical activities and shall report results of statistical analyses to NIEHS committees and workgroups as needed.

4) The contractor shall maintain an awareness of new statistical, mathematical and computational methods and advise the COR regarding possible utilization of these procedures.

5) There shall be frequent interaction between the contractor and the COR or other designated DNTP or DIR researchers, as there are often issues of experimental design, data analysis, and understanding of data analyses that require in-depth discussion. Such interactions usually occur with short notice (less than 24 hours). Moreover, the contractor shall attend various NIEHS meetings identified by the COR which use information provided by the contractor in planning further activities. Therefore, the Contractor shall http://www.niehs.nih.gov/index.cfm http://www.nih.gov/ http://www.niehs.nih.gov/research/atniehs/dir/index.cfm http://ntp.niehs.nih.gov/ http://www.niehs.nih.gov/research/atniehs/dntp/index.cfm maintain an operational office near Research Triangle Park, North Carolina, 27709 so as to provide a one-hour response time.

6) The Contractor shall ensure that all contractor personnel are adequately trained, possess the required credentials, and are otherwise fully qualified and skilled to perform assigned work for this SOW including staying current with developments in the field. The Contractor shall be responsible for all costs associated with travel and training to ensure that staff maintains required certifications and proficiency to perform assigned duties.

7) The contractor Principal Investigator (PI) shall devote a minimum of 75% of their time to this contract.

8) The actual number of studies conducted for which analyses are required may be higher or lower than indicated below.

9) All final reports, software, and other electronic deliverables must be Section 508 compliant.

10) The Government shall provide the Contractor access to the NIEHS and NIH library (http://nihlibrary.nih.gov/researchtools/Pages/default.aspx) facility and electronic resources (http://nihlibrary.nih.gov/researchtools/Pages/default.aspx) (upon proper security clearance). In order to access the NIEHS and NIH library facility and resources, the Contractor staff must obtain NIEHS security badges and adhere to NIH/NIEHS requirements for connecting to the NIEHS information technology resources. Equipment used for library access, such as computers and printers, shall support Internet and Intranet connections, be compatible with NIH computer operating systems, and comply with NIH security policies.

B. Support for the National Toxicology Program (NTP)

1) For NTP studies conducted under Good Laboratory Practices (GLP) guidelines, the contractor shall apply GLP guidelines to the data analyses and reporting.

2) The NTP conducts long-term rodent carcinogenicity studies for which the data typically consist of survival times, body weights, indicators of presence or absence of tumor and non-neoplastic lesions of individual animals (see Reports & Abstracts by Study Types links at http://ntp.niehs.nih.gov/results/pubs/index.html).

For some studies, food or water consumption data are also collected. Approximately 4-6 long-term studies are completed per year. Routine statistical analyses for these studies are automatically generated by the Toxicology Data Management System Enterprise (TDMSE) which, in the near future, will be transferred to, and generated by, the Chemical Effects in Biological Systems (CEBS) database (http://www.niehs.nih.gov/research/resources/databases/cebs/index.cfm). For some studies, however, additional special analyses of these data are required. The contractor shall conduct approximately 15 “special” analyses of tumors or non-neoplastic lesions per year, in consultation with the COR. These special analyses include testing the significance of dose-response trends and of pairwise differences between each dose group and the control group for non-routine situations such as non-standard pooling of dose groups, non-standard pooling of tumor types, or extended evaluations of tissues.

3) Data from short-term rodent toxicity studies typically consist of body weights, organ weights, clinical chemistry, hematology, sperm motility and vaginal cytology evaluations (see links available from http://ntp.niehs.nih.gov/testing/types/cartox/index.html). Data may also include special studies of tissue concentrations, reproductive endpoints, immunotoxicity endpoints, neurobehavioral endpoints.

Approximately 30-40 toxicity and special studies are completed per year and about half of these studies are reproductive, immunologic or neurobehavioral toxicity studies. In consultation with the COR, the contractor shall summarize the data and identify potential outliers for review by DNTP staff. Upon finalization of the data, the contractor shall summarize the final data and test for dose-related trends and pairwise differences from the control group.

4) Approximately 5-8 Modified One-Generation Studies, Reproductive Assessment by Continuous Breeding

(RACB) and/or teratology studies will be conducted per year (see links at http://nihlibrary.nih.gov/researchtools/Pages/default.aspx http://nihlibrary.nih.gov/researchtools/Pages/default.aspx http://ntp.niehs.nih.gov/results/pubs/index.html http://www.niehs.nih.gov/research/resources/databases/cebs/index.cfm http://ntp.niehs.nih.gov/testing/types/cartox/index.html http://ntp.niehs.nih.gov/testing/types/index.html). These multigenerational studies produce data on fertility and fecundity; litter characteristics including, but not limited to, numbers of male and female pups, survival; pup and adult measurements including, but not limited to, body and organ weights, times to vaginal opening and preputial separation, histopathology of reproductive and other organs, nipple counts, estrous cyclicity, ovarian follicle counts, sperm morphology and function. Some studies may also have immunotoxicity endpoints and/or neurotoxicity/behavioral endpoints. In consultation with the COR, the contractor shall conduct statistical tests on these data, including tests for trends with dose, pairwise comparisons with the control group, and correlations between selected endpoints (depending on the material under study). Statistical analyses of these data often require adjustment for litter effects. Because these are complex, time-consuming studies and because intermediate results may influence subsequent steps of the study, it is expected that portions of the data will be finalized at different times.

5) In consultation with the COR, the contractor shall submit a formal historical control report summarizing the historical control data such as tumor rates, growth curves, and survival curves, and where appropriate, fertility rates, developmental indices and malformation rates from NTP studies, including multigenerational studies, approximately every 9 months for each study design (see, for example, http://ntp.niehs.nih.gov/results/dbsearch/historical/index.html). These data are currently stored in the NTP’s CEBS database, maintained under a separate contract.

6) For some compounds under study, the NTP conducts studies of its absorption, distribution, metabolism, and elimination (ADME), toxicokinetics (TK), and/or physiological-based pharmacokinetics (PBPK) (e.g., see Appendix J of Technical Report 546 and/or Appendices M and N of Technical Report 549 at http://ntp.niehs.nih.gov/results/pubs/longterm/reports/longterm.index.html). In consultation with the COR, the contractor shall provide statistical and mathematical modeling support for these ADME, TK and PBPK studies. It is anticipated that approximately 5 chemicals per year shall require modeling.

7) Through the Tox21 program (http://ncats.nih.gov/research/reengineering/tox21/tox21.html and http://ntp.niehs.nih.gov/results/hts/index.html), the NTP conducts high throughput and medium throughput screening assays on hundreds to several thousand chemicals. The high throughput assays include more than one hundred different in vitro assays conducted on approximately 10,000 compounds (http://ntp.niehs.nih.gov/results/hts/index.html). In the near future, medium throughput assays may involve data on zebrafish. In consultation with the COR and NTP staff, the contractor shall provide statistical, mathematical and computational support for high throughput and medium throughput screening studies.

Statistical support includes testing for dose-response associations and comparing responses across assays and across chemicals and simulating data for methods comparisons. Mathematical support includes development of models for extracting information from the data. Computational support includes development of pipelines for efficient data analysis. Analysis methods for these data are evolving rapidly, so the contractor shall stay current with methods for analyses of these data and shall be able to conduct simulation studies to compare various methods.

8) The DNTP occasionally conducts basic research to elucidate responses observed in rodent studies.

Examples of recent research include evaluations of effects of mixtures of chemicals; use of non-neoplastic endpoints to predict eventual cancer; and mechanistic studies; further examples can be found in Bhusari et

al. (2014) Toxicologic Pathology 42:555-564 and Burns et al. (2014) Toxicological Sciences 136:514-526.

In consultation with the COR, the contractor shall provide statistical, mathematical, and computational support to basic research efforts of the DNTP. Support activities include assisting with statistical issues related to experimental design and data analysis, carrying out statistical analyses of data, and developing mathematical models to describe relationships and associations. It is anticipated that there will be approximately 2-3 studies of this type per year requiring statistical support.

9) The DNTP occasionally conducts studies with humans, such as biomonitoring studies, pharmacokinetics studies, product use assessments, etc. In consultation with the COR, the contractor shall provide statistical, mathematical, and computational support for these human studies conducted by the DNTP. Support activities include assisting with statistical issues related to experimental design and data analysis, carrying out statistical analyses of data, and developing mathematical models to describe relationships and http://ntp.niehs.nih.gov/testing/types/index.html http://ntp.niehs.nih.gov/results/dbsearch/historical/index.html http://ntp.niehs.nih.gov/results/pubs/longterm/reports/longterm.index.html http://ncats.nih.gov/research/reengineering/tox21/tox21.html http://ntp.niehs.nih.gov/results/hts/index.html http://ntp.niehs.nih.gov/results/hts/index.html associations. It is anticipated that there will be approximately 2-3 studies of this type per year requiring statistical support.

10) As directed by the COR, the contractor shall write GUI-based programs, typically in Java, that incorporate statistical methods required by the DNTP, which may developed by or with NIEHS statisticians (http://www.niehs.nih.gov/research/resources/software/biostatistics/index.cfm) or may be based on existing methods in the literature (e.g., http://ntp.niehs.nih.gov/help/browsers/metadata/index.html). It is anticipated that approximately 1 method will be required per year.

11) The contractor, in consultation with the COR, shall establish and maintain a procedure whereby each contract laboratory conducting NTP studies can transfer data electronically to the contractor for analysis.

The statistical contractor shall carry out appropriate analyses of data received and, within the specified time frame, provide written electronic reports that summarize the results of the statistical analyses to the NTP at NIEHS. When directed by the COR, the contractor shall also provide written electronic reports to the contract laboratory.

12) It is anticipated that, over the course of this contract, additional study designs may be adopted by the NTP, which may require statistical analysis methods specific to these designs.

C. Support for the DIR

1) Several researchers in the DIR collect data in large-scale surveys on humans, such as Aggarwal et al.

(2014) Arthritis and Rheumatology 66:740-749. In consultation with the COR, the contractor shall provide statistical and computational support to NIEHS’ human studies. These activities include providing advice on statistical aspects of experimental design (including sample size and power) and statistical analysis of data. It is anticipated that there will be 5 of these studies per year requiring statistical support.

2) The DIR conducts basic research at the molecular, cellular, and organismal level. Examples and descriptions of research are available at http://www.niehs.nih.gov/research/atniehs/dir/index.cfm. In consultation with the COR, the contractor shall provide statistical, mathematical, and/or computational support to DIR’s research efforts. These support activities include assisting with statistical issues related to experimental design and data analysis, carrying out statistical analyses of data, and assisting in the development of mathematical models for describing associations or mechanisms. It is anticipated that there will be approximately 2-3 studies of this type per year requiring statistical support.

3) As directed by the COR, the contractor shall write GUI-based programs, typically in Java, that incorporate new statistical methods developed by or with NIEHS statisticians (http://www.niehs.nih.gov/research/resources/software/biostatistics/index.cfm). It is anticipated that approximately 1 new method will be developed each year.

D. 508 Compliance

Requirements for accessibility based on Section 508 of the Rehabilitation Act of 1973 (29 U.S.C. 794d) are determined to be relevant for the Statistical Support contract. Section 508 requires that federal agencies’ electronic and information technology (EIT) is accessible to people with disabilities. The law applies to all Federal agencies when they develop, procure, maintain, or use electronic and information technology. In order to comply with the Section 508 Electronic and Information Technology Accessibility Standards, the contractor shall perform all work required under this contract in compliance with the standards delineated in Code of Federal Regulations (CFR) Title 36 and outlined below.

The following Section 508 accessibility standards apply to the tasks identified in this SOW:

36 CFR Part 1194-Subpart D—“Information, Documentation, and Support”

• 1194.41

The checklists and standards are found at: http://www.section508.gov/. The Contractor shall refer to this website for further guidance.

http://www.niehs.nih.gov/research/resources/software/biostatistics/index.cfm http://ntp.niehs.nih.gov/help/browsers/metadata/index.html http://www.niehs.nih.gov/research/atniehs/dir/index.cfm http://www.niehs.nih.gov/research/resources/software/biostatistics/index.cfm http://www.section508.gov/

E. Reporting Requirements

The following electronic reports are required. Quantities may be somewhat higher or lower than listed. Due dates are specified.

(1) The contractor shall provide to the Contracting Officer and the COR a monthly report describing progress in all areas and problems encountered. (12 per year, by the 10th of each month)

Statistical reports to be provided to the COR:

(2) Statistical reports summarizing data analyses of toxicity, special, and similar studies, 3 weeks following receipt of final raw data. (Approximately 30-40 per year)

(3) Statistical reports summarizing special analyses of tumors or non-neoplastic lesions, 2 weeks following receipt of request and final raw data. (Approximately 15 per year)

(4) Statistical reports summarizing analyses of Developmental and Reproductive Toxicology studies such as

Modified One-Generation, Reproductive Assessment by Continuous Breeding or Teratology studies, 4 weeks following receipt of request and final raw data. (Approximately 5-8 studies per year)

(5) Reports on toxicokinetic models for selected chemicals, 12 weeks after receipt of final raw data.

(Approximately 5 per year)

(6) Historical Control Reports for NTP studies, 2 weeks after request (One for each study design per year).

(7) Reports describing the statistical methods used and providing the data analysis results for individual NTP and

DIR projects, as specified by the COR, 2 weeks after request (number will vary).

Data generated by this contract shall be maintained in electronic media and/or downloaded to appropriate files or formats, as designated by the COR. The contractor shall submit reports in an electronic and/or paper format specified by the COR. All final reports, software, and other electronic deliverables must be Section 508 compliant.

Other Deliverables

• IT Security statement – Due 30 days after contract award

• Section 508 Compliance report – Due annually on the anniversary date of the award

Addresses/phone numbers

Dr. Grace Kissling Project Officer / Contracting Officer’s Representative Biostatistics and Computational Biology Branch National Institute of Environmental Health Sciences P.O. Box 12233, Mail Drop A3-03 Research Triangle Park, NC 27709 Telephone: (919) 541-1756 Email: kissling@niehs.nih.gov

Wanda Holliday Contracting Officer Research Contracts Branch, OM, Office of Acquisitions National Institute of Environmental Health Sciences Research Triangle Park, NC 27709 Telephone: (919) 541-0943 Email: hollida1@niehs.nih.gov mailto:hollida1@niehs.nih.gov