1 - Journal of Clinical Oncology by pengxiang


Special Article submission:
Title: Recommendations for Collection and Handling of Specimens from Group Breast Cancer
Clinical Trials, from Onsite Collection through Shipping to the Central Bank
An international collaboration among BIG and the NCI GBC and Cooperative Groups*

Brian R. Leyland-Jones, M.D., Ph.D.1,
Christine B. Ambrosone, Ph.D.2,
John M.S. Bartlett, Ph.D.3,
Matthew J. C. Ellis, M.B., B.Chir., Ph.D.4,
Rebecca A. Enos, MPH5,
Adekunle Raji B.S.6,
Michael R. Pins, M.D.7, Jo Anne Zujewski, M.D.8,
Stephen M. Hewitt, M.D., Ph.D.9,
John F. Forbes, FRACS, M.S., FRCS, MBBS10,
Mark Abramovitz, Ph.D.11,
Sofia Braga, M.D.12,
Fatima Cardoso, M.D.12, Nadia Harbeck, M.D.13,
Carsten Denkert, M.D.14,
Scott D. Jewell, Ph.D.15
On behalf of the Blood, Formalin Fixed Paraffin Embedded (FFPE), and Fresh/Frozen Tissue
Working Groups of the NCI Cooperative Group Banking Committee, Breast International Group,
and the North American breast cancer Cooperative Groups. (Working group rosters are
provided below.)
1. Emory University School of Medicine, Atlanta, Georgia
2. Roswell Park Cancer Institute, Buffalo, New York / Southwest Oncology Group
3. Endocrine Cancer Group, University of Edinburgh, Edinburgh Cancer Research Centre,
    Scotland, United Kingdom
4. Washington University in St. Louis School of Medicine, St. Louis, Missouri /
5. The EMMES Corporation, Rockville, Maryland
6. Eastern Cooperative Oncology Group Pathology Coordinating Office, Northwestern
    University, Evanston, Illinois
7. Advocate Lutheran General Hospital Department of Pathology, Park Ridge, Illinois (formerly
    of ECOG Pathology Coordinating Office, Northwestern University, Evanston, Illinois)
8. Cancer Therapy Evaluation Program, National Cancer Institute, Rockville, Maryland
9. Tissue Array Research Program (TARP), Laboratory of Pathology, Center for Cancer
    Research, National Cancer Institute, Bethesda, Maryland10- Australian New Zealand Breast
    Cancer Trials Group and University of Newcastle, Newcastle, New South Wales, Australia
10. Australian New Zealand Breast Cancer Trials Group and University of Newcastle,
    Newcastle, New South Wales, Australia
11. VM Institute of Research, Montreal, Quebec, Canada
12. Jules Bordet Institute, Brussels, Belgium
13. Technical University of Munich (Technische Universität München), Munich, Germany;
    German Breast Group (GBG); AGO TRAFO
14. Charité University Hospital, Berlin, Germany/ German Breast Group (GBG), Neu-Isenburg,
15. Cancer and Leukemia Group B Pathology Coordinating Office, Ohio State University,
    Columbus, Ohio / NCI Cooperative Group Banking Committee

                                          Page 1
Work on this document was made possible by support from the Breast Cancer Research
Foundation, the National Cancer Institute (NCI), the Breast International Group, and the NCI-
supported Cooperative Groups.

Corresponding author: Brian Leyland-Jones, MD, PhD, Winship Cancer Institute, Emory
University, 1365-C Clifton Road, Atlanta, GA 30322; Tel: 404-778-5669; Fax: 404-778-5048;

These       guidelines     are     referenced     on     the   World    Wide      Web      at

*Groups contributing to this manuscript:

The Groups of the Breast International Group (BIG) and TRANSBIG
The National Cancer Institute (NCI)
The Groups of the NCI Cooperative Group Banking Committee (GBC), which include the

     The Cooperative Groups of the Breast Cancer Intergroup of North America (TBCI):
          American College of Surgeons Oncology Group (ACOSOG)
          Cancer and Leukemia Group B (CALGB)
          Eastern Cooperative Oncology Group (ECOG)
          North Central Cancer Treatment Group (NCCTG)
          National Cancer Institute of Canada Clinical Trials Group (NCIC CTG)
          Southwest Oncology Group (SWOG)
     National Surgical Adjuvant Breast and Bowel Project (NSABP)
     Radiation Oncology Group (RTOG)
     Gynecologic Oncology Group (GOG)
     Children's Oncology Group (COG)

                                           Page 2

Dr. Martine Piccart-Gebhart; Dr. William C. Wood; Dr. Larry Norton; the Breast Cancer
Research Foundation; specimen banks of BIG, TRANSBIG, and the Cooperative Groups; the
EORTC PathoBiologyGroup; Dr. Manfred Schmitt; the Group Banking Committee (GBC) Best
Practices and Operations Subcommittee; and Leah Kamin.

Blood Working Group Members and Advisors
Brian Leyland-Jones, M.D., Ph.D. (working group chair), Winship Cancer Institute, Emory
University School of Medicine, Atlanta, Georgia
Mark Abramovitz, Ph.D., VM Institute of Research, Montreal, Quebec, Canada
Christine Ambrosone, Ph.D., Roswell Park Cancer Institute, Buffalo, New York
Mark Bouzyk, Ph.D., Human Genetics, Emory University School of Medicine, Atlanta, Georgia
Sofia Braga, M.D., Jules Bordet Institute, Brussels, Belgium
Fatima Cardoso, M.D., Jules Bordet Institute, Brussels, Belgium
Warren Davis, Ph.D., Roswell Park Cancer Institute, Buffalo, New York
Matthew Ellis, M.B., Ph.D., Washington University in St. Louis School of Medicine, St. Louis,
Rebecca Enos, MPH, The EMMES Corporation, Rockville, Maryland
Nadia Harbeck, M.D., Technical University of Munich (Technische Universität München),
    Munich, Germany; German Breast Group (GBG); AGO TRAFO
Daniel F. Hayes, M.D.-University of Michigan Comprehensive Cancer Center, Ann Arbor,
Klaus Pantel, M.D., Ph.D. Universitätsklinikum Hamburg, Hamburg, Germany
Michael Pins, M.D., Advocate Lutheran General Hospital Department of Pathology, Park Ridge,
    Illinois (formerly of ECOG Pathology Coordinating Office, Northwestern University,
    Evanston, Illinois)
Adekunle Raji, B.S., Eastern Cooperative Oncology Group Pathology Coordinating Office,
    Northwestern University, Evanston, Illinois
Federico Rojo, M.D. Vall d’Hebron University Hospital, Barcelona, Spain
Geraldine Thomas, Ph.D., Department of Histopathology, Hammersmith Hospital, Imperial
    College, London, United Kingdom
Jo Anne Zujewski, M.D., Cancer Therapy Evaluation Program, National Cancer Institute,
    Rockville, Maryland
Provided comments:
    Sandra (Brewer) Stakely, M En, and Jaclyn Smith, BS, Nationwide Children’s Hospital,
    Biopathology Center, Columbus, Ohio, Jacqueline M. Lafky. M.S., North Central Cancer
    Treatment Group Biospecimen Resource Manager, Mayo Clinic, Rochester, Minnesota/NCI
    Group Banking Committee

FFPE Working Group Members and Advisors
Adekunle Raji (working group chair), Eastern Cooperative Oncology Group Pathology
    Coordinating Office, Northwestern University, Evanston, Illinois
Mark Abramovitz, Ph.D., VM Institute of Research, Montreal, Quebec, Canada
John M.S. Bartlett, Ph.D., Endocrine Cancer Group, University of Edinburgh, Edinburgh Cancer
    Research Centre, Scotland, United Kingdom
Sofia Braga, M.D. Jules Bordet Institute, Brussels, Belgium
Carsten Denkert, M.D., Charité University Hospital, Berlin, Germany/ German Breast Group
    (GBG), Neu-Isenburg, Germany
Rebecca Enos, MPH, The EMMES Corporation, Rockville, Maryland
Wedad Hanna, M.D., Sunnybrook and Women's College Health Sciences Centre, University of
    Toronto, Toronto, Ontario, Canada

                                         Page 3
Daniel F. Hayes, M.D., University of Michigan Comprehensive Cancer Center, Ann Arbor,
Stephen M. Hewitt, M.D., Ph.D., Tissue Array Research Program (TARP), Laboratory of
     Pathology, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland
Scott Jewell, Ph.D. Cancer and Leukemia Group B Pathology Coordinating Office, Ohio State
     University, Columbus, Ohio / NCI Cooperative Group Banking Committee
Denis Larsimont, M.D., Jules Bordet Institute, Brussels, Belgium
Brian Leyland-Jones, M.D., Ph.D., Winship Cancer Institute, Emory University School of
Medicine, Atlanta, Georgia
Rudolf Maibach, Ph.D., International Breast Cancer Study Group, Coordinating Center, Bern,
Soonmyung Paik, M.D., National Surgical Adjuvant Breast and Bowel Project, Division of
     Pathology, Pittsburgh, Pennsylvania
Michael Pins, M.D., Advocate Lutheran General Hospital Department of Pathology, Park Ridge,
Illinois (formerly of ECOG Pathology Coordinating Office, Northwestern University, Evanston,
Federico Rojo, M.D., Vall d’Hebron University Hospital, Barcelona, Spain
Sheila Taube, Ph.D., Cancer Diagnosis Program, National Cancer Institute, Rockville, Maryland
Marc van de Vijver, M.D., Ph.D., Netherlands Cancer Institute (Nederlands Kanker Instituut),
     Amsterdam, The Netherlands
Giuseppe Viale, M.D. - University of Milan (Università degli Studi di Milano), Milan, Italy
Jo Anne Zujewski, M.D., Cancer Therapy Evaluation Program, National Cancer Institute,
     Rockville, Maryland
Advised at meeting of working groups: Gunter von Minckwitz, M.D., Ph.D., German Breast

Fresh/Frozen Tissue Working Group Members and Advisors
Matthew Ellis, M.B., Ph.D. (working group chair), Washington University in St. Louis School of
    Medicine, St. Louis, Missouri / CALGB/ACOSOG
Mark Abramovitz, Ph.D., VM Institute of Research, Montreal, Quebec, Canada
Sofia Braga, M.D. Jules Bordet Institute, Brussels, Belgium
Rebecca Enos, MPH, The EMMES Corporation, Rockville, Maryland
John Forbes, FRACS, M.S., FRCS, MBBS, Australian New Zealand Breast Cancer Trials Group
    and University of Newcastle, Newcastle, New South Wales, Australia
Nadia Harbeck, M.D., Technical University of Munich (Technische Universität München),
    Munich, Germany; German Breast Group (GBG); AGO TRAFO
Stephen M. Hewitt, M.D., Ph.D., Tissue Array Research Program (TARP), Laboratory of
    Pathology, Center for Cancer Research, National Cancer Institute, Bethesda, Maryland
Brian Leyland-Jones, M.D., Ph.D., Winship Cancer Institute, Emory University School of
    Medicine, Atlanta, Georgia
Stella Mook, M.D., Netherlands Cancer Institute (Nederlands Kanker Instituut), Amsterdam, The
John Olson, M.D., Ph.D., Duke University, Durham, North Carolina, ACOSOG
Federico Rojo, M.D. Vall d’Hebron University Hospital, Barcelona, Spain
Arun Seth, Ph.D., Sunnybrook Health Sciences Centre, University of Toronto, Toronto, Ontario,
Baljit Singh, M.D., New York University, New York, New York
Laura van ‘t Veer, Ph.D., Netherlands Cancer Institute (Nederlands Kanker Instituut),
    Amsterdam, The Netherlands
Mark Watson, M.D., Ph.D., Washington University in St. Louis School of Medicine, St. Louis,
Jo Anne Zujewski, M.D., Cancer Therapy Evaluation Program, National Cancer Institute,
    Rockville, Maryland
                                         Page 4
Provided comments: Sandra (Brewer) Stakely, M En, and Jaclyn Smith, BS, Nationwide
   Children’s Hospital, Biopathology Center, Columbus, Ohio

                                    Page 5
Minimum Facility Equipment and Space Requirements

Submitting institutions should have specimen-processing equipment, particularly centrifuges

and -20°C or colder storage. If specimens cannot be stored in a -80°C freezer, then -20°C

freezers can be used, followed by shipment of the specimen within 24 hours of collection to the

central bank (see “Shipping” below). This especially may be the case for outlying and small-

hospital collection sites. If a site lacks processing ability, whole blood intended for serum/plasma

processing must be shipped priority overnight to the clinical trial repository with which it is

affiliated. The clinical trial repository will need to process, aliquot, and freeze these specimens

on the day of receipt, especially if specimens are shipped on Friday for Saturday delivery.

If an institution cannot meet the requirements of a clinical trial’s protocol for onsite specimen

collection, freezing, or processing, then that site should not participate in the clinical trial. When

developing a trial, clinical trial groups need to consider whether certain participating sites who

cannot meet certain laboratory standards should be allowed to enroll patients. Such

consideration is especially important if the biospecimens collected are vitally important to the

research or purpose of the trial. Allowing multiple acceptable standards for specimen

preparation in the same trial should only occur if documentation/annotation of specimen

processing clearly characterizes the handling of the specimen, including the unprocessed

period. Proposed requirements for such annotation are provided in this document.

Central repositories should follow the recommendations of the NCI Best Practices for

Biorepositories, which, at a minimum, include -80°C freezers with alarm systems to monitor

freezers daily and notify repository personnel if malfunction occurs (1).

The three working groups stressed the importance of banking biospecimens for yet to be

developed correlative research. Often, excess quantities of biospecimen components are

collected in most trials, to include serum, plasma, DNA, and tissues that allow for future

                                             Page 6
research, in addition to studies already embedded in clinical trial protocols. These biospecimens

are specifically protected for future research studies over and above biomarker directed patient

care. Therefore, funding should be allocated to maximize laboratory and repository space in the

Group banks in order to accommodate the processing and storage of research specimens

recommended in these guidelines.

Specimen collection kits

Pre-assembled kits sent by the central group bank to sites can vastly expedite as well as

standardize the collection of good-quality specimens from sites, reduce the variability in

collection procedures, facilitate onsite processing (such as aliquoting), and increase compliance

with the protocol-stipulated collection. Group banks should use pre-assembled collection kits to

the broadest extent possible.

Kits should contain easy-to-understand, simple instructions for processing and shipping, and be,

for the most part, self-explanatory. They should contain the necessary freezing tubes, blood

draw tubes and needles, disposable transfer pipettes, and other supplies to facilitate onsite

specimen acquisition and processing (e.g., aliquoting and freezing), as well as shipping

materials for sending the specimen to the central repository. The exact contents of each kit will

be protocol-specific, although they should adhere to the recommendations provided above.

Once a local site returns a kit to the central bank, the central bank should send that site a

replacement kit, to ensure that the site will have a kit available for future use (see “Shipping”

below for further information on specimen shipping).

Pathologist Support for Submitting Research Specimens

A great need exists to increase support for pathologists who carry the major workload

associated with submitting research tissue specimens. Recommendations for increasing

                                           Page 7
pathologist support for sending research specimens to the central Group repository include the


   1) Increase pathologist involvement in, and awareness of, clinical research. This is

       paramount. Involvement of a pathologist on the steering committee for a clinical trial,

       particularly clinical trials involving translational research, is essential, since it will inform

       the trial management group of the logistics required for tissue collection. Pathologist

       involvement should help to raise the profile of clinical trials among pathologists,

       particularly through pathology specialist meetings.

   2) Ensure that money allotted to pathology for a clinical trial actually reaches pathologists.

       Although funds may be targeted towards pathology efforts on a clinical trial in a variety of

       ways (e.g., via internal arrangement at the institution itself, as part of per-patient

       reimbursement, etc.); these funds may not actually reach pathologists.

   3) Where appropriate, consider (as a prospective recommendation) re-apportioning Group

       resources to allow increased reimbursement of pathologists. Reimbursement must be

       commensurate with effort. The 2007 NCI Best Practices Guidelines for Repositories (1),

       noted above, recommend a range of $50 - $250 US dollars. Although it was recognized

       that different international cooperative group and research consortia required different

       levels of support and reimbursement. It was regarded as critically important that the

       pathology service was adequately reimbursed to compensate for this unique


   4) Biorepositories must respect the local pathologist desire to return tissue blocks when

       needed for patient care. At the same time, clinical research institutions internationally are

       recognizing the importance of upfront informed consent that permits analysis of the

       patient’s biospecimens for research biomarker studies. For example, in the US, clinical

       cooperative groups can maintain linkage of the patient to the biospecimens. While

       research derivatives (tissue sections) are not permitted to have patient linking

       information when used by the investigator, the cooperative group data centers can

                                             Page 8
       connect patients to institutions for the timely return of the diagnostic block if needed for

       patient care, pointing out that, in almost all situations, sending a block back to an

       institution should not be required.

   5) Increase pathologist awareness of existing studies that demonstrate the impact of

       handling and processing procedures on specimen quality.

   6) Emphasize reasons why research block submission is essential.

   7) Again, self-explanatory kits for specimen submission, with brief instructions, can greatly

       facilitate pathologist submission of research specimens.

Central Banking of and Access to Research Specimens

Specimens from a clinical trial, particularly FFPE tissues, must be consolidated in a central bank

independent from industry. These cooperative Groups biorepositories provide a significant

challenge internationally in terms of long-term maintenance, funding, and governing policies for

access and patient public protection. Other significant issues include intellectual property,

appropriate sharing and access of resources at local, national, and international levels.

Centralization of specimens will greatly facilitate access to specimens by investigators with

meritorious proposals for specimen-based studies. Each central bank should have a disaster

policy in place with details for carrying out this plan provided in the manual of operation; many

biorepositories are building separate physical facilities, which duplicate the original specimen

collection housed several miles apart.

The "honest broker" and "safe harbor" systems provide a high level of comfort to member sites,

institutional pathologists, and patients. These intrinsic measures ensure confidentiality and

privacy while providing researchers with necessary information and high-quality samples.

Independence from industry is important, since efforts by industry to claim intellectual property

or other rights to research specimens may compromise research.

                                             Page 9
Intellectual property (IP) issues related to clinical trial tissue banks must be considered. Most

recently, intellectual property is being defined at the beginning of the clinical trial and

subsequent use of the biospecimens in research. A Material Transfer Agreement (MTA) that

clearly identifies ownership of the biological material and its associated IP is an important

component of any biorepository. Appropriate sharing of royalty streams between researcher,

university, cooperative groups, and sponsors are now being negotiated up front as parts of

standard agreements.

Researchers should be able to submit proposals for use of clinical trial specimens (which would

include tissue microarray sections). The limited quantity and high value of clinical trial

specimens warrant their careful and triaged use. A committee should be in place, or

established, using pre-specified policies, to vet proposals in order to prioritize specimen usage

appropriately, and to screen out proposals that are without merit or that contain serious


International trials:

A central tissue bank should be established for each trial, with custodianship vested in the trial’s

steering group. An educational push is needed among international clinical trial Groups for sites

to send in biological material for research, and then to centralize those biological materials.

Increasing recognition must be paid to the two-thirds of the global population living within Asian

communities and the importance of this critical resource in terms human biomarker correlative


North American trials:

In North America, specimens for a given trial should be centralized at the level of the Group

leading the trial; this also pertains to specimens from intergroup trials. Retrospective as well as

prospective consolidation of intergroup specimens is important to enable research to be carried

                                           Page 10
out on those specimens. Especially, intergroup specimens from closed trials that are still stored

at various Groups or at various sites need to be consolidated in the lead Group’s bank.

FFPE tissue

FFPE tissue is most often used for immunohistochemistry, in-situ hybridization, or extracting

DNA and RNA. It is essential that a common baseline procedure be in place for the collection

and handling of FFPE tissues in breast cancer clinical trials. A large proportion of FFPE tissue

will be used for diagnosis as well as for translational research.

Processing requirements must be congruent with current pathology practices and guidelines

(e.g., College of American Pathologists). Furthermore, a guiding principle for stewardship of

FFPE tissues should be the accommodation of different analytical platforms.

Mandatory FFPE tissue submission for research

These guidelines promote the ethical use of consented human tissues and recognize the major

contribution to the furtherance of science from patients who freely consent to donate their tissue

for research. Withholding these tissues without appropriate cause, following patient consent,

frustrates patients’ wishes and is potentially equally as damaging as failure to obtain consent.

If a patient consents to the use of her/his tissue for research, the appropriate specimens must

be submitted as outlined in the trial protocol, provided this does not compromise future local

diagnostic procedures. While recognizing that the preservation of such material is a primary

requirement for the clinical pathologist of record, it is also recognized that cases where this

requirement precludes submission of material for research will be rare. Indeed, except for very

small tumors, current practice involves disposal of large amounts of unused tumor material that

could be used for research.

                                            Page 11
Failure to submit these specimens will be considered a violation of the Cooperative Research

Agreement as well as a deviation of protocol, which can ultimately have an impact on the

institutional Federal-Wide Assurance (FWA) and member access to Cooperative Group breast

cancer specimens.

Consistent or continued refusal of a site to submit consented research specimens to the trial’s

central bank should be considered reason for exclusion of that site from Group membership,

support, and access to specimens by its researchers. This latter measure recognizes that failure

to cooperate in joint ventures may also compromise the quality of the science.

The trial’s central bank should retain the submitted blocks, but will return a block to the site in a

timely manner if it is needed for patient care (see item #4 under “Pathologist Support for

Submitting Research Specimens”).

The following includes the optimal range of samples that should be submitted either within one

month if in North America and within eight weeks maximum if outside of North America, or

according to the protocol (since some protocols ask for shipment in batches to reduce costs).

Sites outside of North America may obtain special provision for submission of tissue to North

American central banks in order to save cost; sites interested in this option should check with

the lead Group office to discuss this possibility.

   1. Block of primary tumor and block of adjacent non-tumor tissue for permanent storage at

       the central repository. The tumor block should be a sample of the patient’s tumor

       obtained from the surgical resection specimen that is used in the final diagnosis and

       clinical staging. The block should ideally contain at least 5 mm3 of invasive tumor,

       preferably taken from the periphery of the tumor in order to avoid central tumor necrosis

       and to include the most biological relevant area of the tumor. In the case of small tumors

       it is clear that a significant level of judgment to submit specimens for research must rest

                                             Page 12
   with the pathologist of clinical record. Note: When choosing a research block, the area of

   tumor where there has been a previous needle biopsy before surgery should be avoided,

   since inflammatory material and RNA/protein changes near this site can greatly interfere

   with the results of specimen-based analyses.

2. If submission of primary tumor research block(s) is not possible, the diagnostic tissue

   block should be submitted to the trial’s central bank for core extraction for tissue

   microarrays (TMAs), which will be permanently stored at the central bank. An additional

   core for RNA/DNA extraction may also be taken. Note: Details on TMA construction and

   TMA quality assurance can be found in “Tissue Microarray (TMA) Construction and

   Quality Assurance by the Central Bank”, below.

3. If neither of the above can be submitted, a letter should accompany blocks that are

   requested to be returned, that clearly explain why permanent block submission of the

   individual case to the central bank is not possible.

4. If local and/or distant recurrence tumor is resected, and a tumor block for research can

   be made available from this recurrence, it should be submitted to the trial’s central bank

   for permanent storage there. If an axillary node dissection contains tumor, a lymph node

   tumor block should also be submitted. If no research block is available from the

   recurrence or lymph node resection, but a diagnostic block is, this should be submitted

   for coring as described in item 2 above.

5. The trial’s pathology material submission form should be submitted with the specimens

   along with other documentation required by the protocol (this form is distinct from the

   pathology or surgical reports, which contain identifying information).

6. For recommendations on core needle biopsy collection in preoperative studies, please

   see “Fresh/frozen tissue”, below. It should be noted that core needle biopsy specimens

   are actually the most representative tumor specimens, since changes induced in the

   tumor tissue by the core needle biopsy will still be present in the surgical specimen and

   thus may severely affect results obtained from the surgical specimen.

                                       Page 13
The site must notify the trial’s central bank when a sample has been shipped (by fax, email, or

other electronic tracking system). For information on shipping of FFPE samples, see “Shipping”,


While unstained FFPE tissue sections may be the only source of tissue from a patient’s case

that can be provided by an institution, submission of unstained tissue sections should be

discouraged and the last option for the contribution of tissue to meet a protocol. Tissue sections

are often used in IHC or in-situ hybridization for the identification of tumor-associated

biomarkers. However, when tissue sections have been cut for longer than days to months from

the paraffin block, loss of the detection of biomarkers can occur (2). Since storage of tissues

and thus tissue sections in Group biorepositories can exist for several months to years, it is

unwise to collect tissue sections as the main biospecimen. Submission of unstained sections

should also be determined on a case-by-case basis and not as a routine policy of submission

unless allowed as defined in the clinical trial protocol.

Retainment of Paraffin Blocks at Submitting Institutions

In the US, the College of American Pathologist (CAP) guidelines mandate time periods on the

requirement to maintain diagnostic paraffin blocks and slides. Many community hospitals adhere

to these policies while most academic pathology departments greatly exceed these

recommendations for the interest of academic research. Therefore it is important to acquire

relevant research material from community hospitals prior to the end of these guidelines.

However, there is not any guideline for the destruction of the research material in the Group

biorepositories after the guidelines from CAP.

                                             Page 14
Fixation and paraffin infiltration

We strongly recommend that one fixative and one buffer type should be used across Group and

multi-center trials. All tissue samples should be fixed in 10% neutral phosphate-buffered

formalin (i.e., 3.7% formaldehyde), pH 7.

Resection specimens should be grossly dissected (macrosectioned) prior to fixation, to ensure

adequate penetration of the fixative. It is essential that surgical margins are appropriately

marked and that these steps are carried out by a pathologist or their designate. Different blades

should be used when dissecting normal tissue vs. tumor tissue.

The time that elapses from resection to dissection and formalin fixation (the warm ischemia

time) should be minimized, and typically should be no longer than 4 hours. Placing the

specimen in fixative without dissection for overnight or longer is NOT adequate.

Regarding duration of fixation itself, although under-fixation is typically more of a risk than over-

fixation, care should also be taken to avoid prolonged fixation. The following are recommended

acceptable ranges of duration of fixation for different types of FFPE specimens:

      Biopsies (core, needle, and skin biopsies): 8-24 hours

      Excision specimens (e.g., mastectomy, lumpectomy): 12-36 hours

      Tissue sections (0.25-1.0 grams): Overnight-24 hours in formalin (36 hours the absolute


      Weekend specimens: Although fixing and shipping of specimens over the weekend (i.e.,

       Friday-Monday) is discouraged, if mastectomy or lumpectomy specimens must be left in

       formalin over the weekend, they should be “blocked” and placed in a large amount of

       formalin in closed containers.

                                            Page 15
If the collecting center is not associated with a pathology group and does not have access to a

tissue processor or cannot arrange to work with a pathology group that has a tissue processor,

then the specimen should be either 1) shipped in formalin on the day of collection for next-day

delivery to the central bank, or 2) fixed in formalin for 24 hours and then transferred to 70%

ethanol, then shipped to the central bank within a few days.

The following are recommended ranges of time from formalin to paraffin on the processor

(variable according to the specific manufacturer):

      5-8 hrs for biopsies

      6-14 hrs for other specimens (time depending on the instrumentation and tissue size)

Completion of the processes of dehydration with alcohols, clearing with xylene, and

impregnation with paraffin is important. Some findings have suggested that extended processor

times may result in higher-quality analytes, although which step should be prolonged has not yet

been determined (Stephen Hewitt, personal communication, 2006). Use of low-melt paraffins

has been recommended. Contamination with beeswax should be avoided (3).

The following are fixation “Don’ts”:

      Avoid using fixatives for multiple fixation procedures.

      Use of alternative fixatives is discouraged. Assays developed for formalin-fixed

       specimens may not perform as specified with alternative fixatives.

      Additives to formalin that purportedly improve nucleic acid recovery may compromise

       proteins and end-assays.

      DO NOT use mercuric chloride as a fixative: it destroys DNA. B-5 is a fixative that

       contains mercuric chloride and formalin. Bouin’s solution also contains acids, and should

       be avoided.

                                           Page 16
      Do not place undissected specimen in fixative for overnight or longer; the specimen

       should be dissected.

Note: Whereas the tissue processor processes tissue fixed in formalin and infiltrates the tissue

with paraffin, the paraffin-embedding station is a separate piece of equipment that aids the

technologist in placing the tissue into a paraffin block so that it can then be cut.


Re-embedding should only be performed in the rare situation where a block is damaged,

contains artifacts (air bubbles, etc.), or when re-embedding enhances access to small residual

amounts of tissue (e.g., thin blocks, oblique block face, etc.). If a tissue has to be re-processed

at a central bank, the originating hospital should be informed of this.

Storage of FFPE tissue

FFPE blocks should be stored at temperatures below 80ºF (below 26ºC) in an “office-like”

environment – i.e., a controlled-temperature environment with room temperature typical for an

office, and protected from excessive heat (>28C), humidity (>70%), and dryness (<30%

humidity). FFPE tissue should not be stored in basements (danger of water) or warehouses

(danger of insects). Light exposure for FFPE tissue is a key problem and should be diminished.

Vacuum packing of FFPE blocks is recommended. Simple “seal-a-meal” units are affordable

and easy to use. If stored in a refrigerator, avoid cardboard containers and use corrugated

plastic containers. Cardboard boxes in cold rooms/refrigerators will sooner or later soften and

grow mildew.

Storage of unstained FFPE slides (whether from a single block or from a TMA block) is

problematic, due to antigen loss, and should be discouraged (2, 4). Attempts to preserve

antigenicity, e.g. by refrigeration of unstained slides, has not been shown by itself to be of

                                             Page 17
benefit (4). Biomarker analyses may best be carried out on freshly cut FFPE sections. (See

caveat regarding submission of unstained FFPE sections at the end of the “Mandatory FFPE

Tissue Submission for Research” section, above.)

Annotation of Laboratory Methods for FFPE samples

Annotation of FFPE samples may vary from trial to trial, and from bank to bank. We do

recommend, however, that fixative type, buffer type, and time from resection to dissection and

formalin fixation should be reported and recorded in the central bank’s database.

Select quality control measures for FFPE tissues that would be performed by the central bank

can be found in the “Quality Assurance of FFPE Tissue by the Central Bank” section below.

Blood and Serum

Recommended blood collection for clinical trials research

Unless otherwise required, a standard maximum volume of blood for each blood draw is 20 mL.

The following peripheral blood and blood component samples are recommended for collection

from Group clinical trials in breast cancer:

      Whole blood for germline DNA, preferably, although not necessarily, at baseline. DNA

       from whole blood can be used in pharmacogenomic studies.

      Serial collection of serum or plasma. Serum that has been processed appropriately can

       yield protein molecules suitable for proteomic analysis. Although serum is more

       traditionally used for biomarker analysis, plasma may actually be the preferred specimen

       for proteomics. Note: It is understood that not all patient accrual sites, particularly those

       in the community, are equipped or staffed to properly process serum or plasma, which

       should be processed quickly in accordance with the clinical trial protocol and to meet

       quality research standards. Therefore, if accrual sites cannot meet high standards and if

       the protocol offers an alternative to the processing of blood specimens, at minimum,

                                               Page 18
       documentation to determine the unprocessed period must be recorded (see “Required

       Database Annotation for Blood Collection”, below). If the trial protocol does not offer an

       alternative, the accrual site should not be eligible to enroll patients into the trial.

Tubes for blood collection

Standard non-breakable (i.e., plastic) blood collection tubes from various suppliers are

acceptable. It is encouraged that the protocols define specific blood collection tubes, even if

routine or common.

For whole, anticoagulated blood, EDTA is the most preferred agent for plasma and isolation of

peripheral blood mononuclear cells (PBMC) / germline DNA. Again, it should be emphasized

that plastic tubes should be used. PAXgene is also acceptable for collection of whole blood for

germline DNA.

Serum is often collected in clotting tubes with and without serum separating material. After

collection of the blood, allow the blood to clot at room temperature for 20-30 minutes. After the

clot has formed, centrifuge at 2,500 rpm for 15 min at 4°C. The serum can then be removed

from the blood collection tube and transferred (aliquoted) into 1 or 2 mL cryovials for storage at -

80°C, prior to shipment in dry ice.

Plasma can be collected using the BD P100 kit, which contains tubes with both EDTA and

protease inhibitors. The protease inhibitors help to protect and preserve plasma proteins. After

collection of the blood, samples should be immediately chilled in ice water, and then spun at

2,500 rpm for 15 min at 4°C. The plasma can then be removed from the blood collection tube

and transferred (aliquoted) into 1 or 2 mL cryovials for storage at -80°C, prior to shipment in dry


                                             Page 19
Use of heparin (green-top tubes) is generally discouraged, due to its deleterious effect on white

blood cells. Green-top tubes can be used for cytokine measurement, however. For cytokine

measurement, blood must not be collected into glass tubes or vacutainers, but rather into sterile

plastic tubes containing pure heparin.

Time points for serial serum or plasma collection

Serial collection of serum or plasma is strongly recommended because serial samples are

critical to the correlative analysis of markers of response and/or outcome, especially given that

some important markers may only be significant temporally.

To the extent possible, blood draws for serial serum or plasma for research should be

performed at the same time as routine draws or other routine follow-up. All studies should

accrue baseline specimens prior to treatment, followed by collection at intervals to analyze the

interactions of the therapy.

The following time points for blood draws for serum or plasma are recommended:

   Preoperative studies: At baseline; at 2-4 weeks on treatment; at 12-16 weeks after start of

   treatment (usually at time of surgery) or at change to a second regimen of preoperative

   therapy; at time of surgery if the third time point was at change to a second preoperative

   regimen; and at breast cancer event. It is understood that actual time points on preoperative

   studies will be study-specific. It should be noted, however, that changes in marker levels

   have been detected as early as two weeks on treatment, hence the recommendation for the

   second time point above.

   Adjuvant studies: At baseline; at a single time point during the study; at end of therapy; at a

   time point approximately one year after end of treatment; and at breast cancer event.

                                          Page 20
   Metastatic studies: At baseline; at two additional subsequent time points; and at breast

   cancer event or tumor progression (this may be one of the additional time points subsequent

   to baseline).

Onsite processing of blood components

Anticoagulated whole blood for germline DNA extraction should be shipped fresh (unfrozen), on

a cold pack, priority overnight service, to the central laboratory/bank on the same day as

phlebotomy. (See “Shipping” below.) This is because the procedures for the isolation of DNA or

buffy coat of white blood cells from whole blood are best handled in the central laboratory/bank.

To minimize degradation, serum and plasma specimens should be processed, aliquoted, and

frozen onsite at the submitting institution. Frozen aliquots can then be scheduled for shipment

on dry ice shortly thereafter to the central laboratory/bank (see “Storage of Blood and Blood

Components” and “Shipping” below).

If shipment cannot be made immediately – for instance in trials where specimens are shipped

only periodically to reduce costs – aliquots can be stored in a -80°C freezer on-site, and then

batches sent to the central laboratory/bank.

Separation into serum or plasma and placement of the serum or plasma into frozen storage

should occur within a recommended two hours of the blood draw, and sooner whenever

possible. Processing times and centrifugation rate and duration should adhere to the

standardized laboratory SOP and should be clearly defined in the study protocol. Protocols

should also describe the appropriate storage vial(s), quantity of fluid per vial, freezing, short-

term storage conditions, and shipping methods. Guidance on these points is offered in this


                                           Page 21
DO NOT freeze specimens before centrifugation. Otherwise, hemolysis of the red blood cells

will occur and render the serum or plasma unusable.

Centrifugation should occur onsite, and separation of serum at room temperature should occur

prior to centrifugation (generally, normal clotting times allow clot formation and serum

separation within 20-30 minutes at room temperature. Blood clot retraction varies with patient


If serum (after clotting) or plasma cannot be processed immediately, the blood should be

refrigerated at 4ºC. Delays of greater than 18 hours should be avoided.

Serum or plasma samples should be aliquoted into appropriate non-breakable cryovials and

frozen immediately after processing, at -80oC.

Methods for aliquoting are based on if the specimen(s) are processed onsite at the submitting

institution or if specimens are first shipped to a central bank within 24 hours.

Although aliquoting onsite is preferable, it does require attention to detail by multiple

laboratories and is therefore more demanding for the clinical investigators. If processing is not

conducted with adherence to the protocol by the site, additional variables can be introduced and

uniformity across the trial reduced. For example, as indicated below, each patient sample must

be aliquoted into multiple vials so that a series of samples from each blood draw can be banked

to prevent thawing and aliquoting by the bank prior to use in research. This will require

increased time, effort, and materials at the submitting institution.

Approximately one-half the volume of the whole blood (tube volume) will yield serum or plasma.

Therefore, a 10 mL quantity of blood will yield about 4 to 5 mL of serum or plasma.

                                            Page 22
Using a disposable transfer pipette (DO NOT pour), remove the upper clear layer and aliquot

0.5 or 1.0-1.5 mL serum or plasma into 1.0 or 2.0 mL polypropylene cryovials, respectively.

During the pipetting, take care not to disturb the clot (serum) or the buffy coat from plasma. If

buffy coat (cloudy middle layer in plasma collection tube) and/or red blood cells (bottom layer in

plasma collection tube) are being collected, they may also be transferred and aliquoted at this

time. Discard layers not being collected and aliquoted into biohazard waste.

If multiple tubes are used in the same blood draw for serum or plasma, the resulting serum or

plasma may first be gently combined and mixed before aliquoting.

   Note: Pharmacokinetic and pharmacodynamic studies (as opposed to biomarker studies)

   are often very sensitive to sources of light. If the study drug is photosensitive, cover samples

   with foil, during all collection, processing, and shipping steps. This includes covering

   transparent centrifuge lids with foil, and minimizing exposure of the sample to light in other

   ways (closing blinds, switching off artificial light, and so forth).

Freeze the serum or plasma immediately and store at -80C until it can be shipped to the central

Group bank in dry ice.

Storage of Blood and Blood Components:

Short-term storage of whole blood (before shipping): Fresh whole blood in (plastic) EDTA tubes

should be stored at 4ºC and shipped priority overnight service to the central bank on a cold pack

on the same day as the blood draw if possible. Whole blood should NOT be frozen, since this

temperature results in reduced DNA yield from blood.

Short-term storage of serum and plasma (before shipping): Serum (other than PAXgene) and

plasma samples should be stored at -80C until they can be shipped (in dry ice) to the central

                                             Page 23
Group bank. If a site lacks -80C freezing capability, aliquots should be frozen using a standard

-20oC freezer and then shipped, packed in dry ice, to the central bank within 24 hours if

possible. Friday draws should try to be avoided if possible; however, aliquots can be stored at -

20C on the weekend and then shipped on Monday. PAXgene tubes should be refrigerated

between 2°C and 8°C prior to shipping. Again, the time a serum/plasma sample (other than

PAXgene) spends at warmer (i.e., -20C to -60C) temperatures should be minimized.

Long-term storage: -80°C (or colder, e.g., in liquid nitrogen) is the recommended temperature

for storing blood and its subcomponents, including serum, plasma, RNA, and peripheral blood

mononuclear cells. DNA should be stored optimally between-20ºC and -80ºC. If blood is to be

frozen, it should be aliquoted into plastic centrifuge tubes and stored at -80°C. Whole blood

should NOT be stored (or handled or transported) at -20ºC (see “Notes on Downstream

Processing” below). For aliquoted serum, either vapor or liquid phase nitrogen is an alternative

to -80C. -80°C is also the recommended temperature for long-term storage of PAXgene


For further information on long-term storage, please see the recommendations on storage

conditions contained in the “NCI Best Practices for Biospecimens Resources” (June 2007),

under Section B.1.4, “Biospecimen Storage”, found on pages 4-5 of that document (1).

When possible, specimen collection and shipping kits should be provided to the submitting site

by the central bank. Kits should contain the supplies necessary for the blood draw,

serum/plasma processing, and specimen shipping, including the following:

      Return shipping box and other shipping materials for sending specimens to the central


      Blood draw tubes and needles (especially for PAXgene or specialty tubes)

                                          Page 24
      Extra blood-draw tubes and needles (optional)

      Labels and bar codes for the blood draw tubes

      Alcohol pad for the blood draw

      Transfer disposable pipettes for aliquoting

      Cryovials for containing the aliquots

      Cold-resistant labels or ink and bar codes for the cryovials

      Tracking sheet/laboratory requisition form, with appropriate bar code label

      Biohazard materials and information

      Instructions for processing and shipping

Note: If a kit is not available for a protocol that uses PAXgene and a blood draw is imminent,

drawing into an EDTA tube is acceptable.

Database Annotation for Blood Specimens

Annotation of each and every specimen is a critical determinant in the interpretation of any data

emanating from that specimen.

Therefore, the following information should be collected on each blood draw, and recorded in

the trial database. (See “Labeling” below for details on vial labeling.)

Required database annotation for blood collection:

      Study name or ID

      Patient study ID

      Unique laboratory ID

      Time point of collection (e.g., post-chemo, local recurrence, etc.)

      Additives (coagulants, anti-coagulants, gel, beads, coated/sprayed tube, etc.)

      Volume of blood or blood component collected

                                            Page 25
      Numbers and types of tubes or aliquots collected

      Consent level (e.g., if consent for genetic testing is included)

      Any variations from the protocol (e.g., problems with the draw, if less than the required

       amount was obtained, etc.)

      Date and time of blood draw

      Date and time into processing lab

      Date and time blood processed (e.g., spun, aliquoted)

      If a non-refrigerated centrifuge was used

      Date and time specific sample was transferred to freezing conditions

      Temperature of storage

      If a freeze-thaw(s) occurred (site or central bank)

      Date and time specimen was sent to central bank

      For central bank: Date and time received

      For central bank: Condition of specimen

      For central bank: Where sample is stored (rack number, row, position)

Recommended database annotation for blood collection (not required, although helpful):

      Date and time last was meal eaten (important for future proteomics)

      Date and time last aspirin was taken (important for future proteomics)

      If blood transfusion was received in previous 6 months (important for future proteomics)

      Hemolysis (note degree of hemolysis apparent). Note: If hemolysis is significant, another

       sample should be drawn

      Lipemia

      Date of last menstrual cycle, if relevant

      Name of scientist who processed the blood

      Name of person who drew the blood

                                            Page 26
      Name of person sending specimen to central bank

      For central bank: Name of person receiving specimen


Freeze-thawing should NOT occur. Optimally, serum and plasma should be frozen once after

processing and thawed only once, by the researcher at the time of testing (6). If freeze-thaw has

occurred, the freeze-thaw date should be documented.

Circulating Tumor Cells

The field of circulating tumor cells (CTCs) is yet nascent: methods are largely unpiloted and not

yet validated. Therefore, the CTC field is not yet ready for a guideline on collection and

processing. The field remains a research enterprise.

Fresh/frozen tissue

Flexibility should be permitted in how fresh/frozen tissue is collected, as long as the general

principles of rapid collection and freezing (or transfer to RNAlater®), as well as seamless

integration with standard pathologic assessment, are respected. Adopted methods should also

provide evidence of careful quality assurance and should demonstrate feasibility in a variety of

practice settings.

Some guiding principles for fresh/frozen tissue collection include the following (also see Table I,

“Do’s and Don’ts for fresh/frozen tissue collection”):

      Preoperative and surgical frozen specimens should be frozen using the same freezing


      For any approach using frozen specimens, THE SPECIMEN SHOULD NOT BE


      Different techniques and media may be used to snap-freeze fresh specimens (see Table

       II, “Options for Snap-freezing”)

                                            Page 27
   Controlled snap-freezing in cooled isopentane or with a heat extractor is recommended

    over simple dry-ice freezing

   Snap-frozen specimens should be placed in an appropriate container (e.g., cryovial or

    cassette) and transferred to -80°C freezing conditions or colder (e.g., liquid nitrogen) for


   If optimal cutting temperature (OCT) freezing medium is used, each core should be

    separately embedded in the OCT. Specimen(s) are sufficiently frozen when the OCT has

    become completely white.

   OCT is especially useful for preserving histology, and may improve RNA recovery. A

    frozen section resulting from an OCT-embedded specimen will provide important

    information on the presence and quantity of tumor. Moreover, OCT prevents the tissue

    from desiccation and crumbling and also acts as an insulator from thermal change and

    limits ice crystal formation.

   If RNA/DNA-based analyses are planned, use of OCT is recommended. RNAlater® is

    appropriate for RNA studies as an alternative to snap-freezing, although OCT is optimal

    for preserving morphology. RNAlater® may improve the quality of RNA after long

    storage at -80ºC (7).

   If proteomic analyses are planned, use of either OCT or RNAlater® is NOT

    recommended. Frozen samples planned for proteomic analysis should be stored in liquid


   If RNAlater® is used, the tissue should be sliced into slices of no more than 2 mm.

   Placing the specimen in a tube first before snap-freezing may be optimal; however,

    cryomolds are also commonly used.

                                        Page 28
Table I. “Do’s and Don’ts” for fresh/frozen tissue collection

         Do not directly place the specimen in liquid nitrogen
         Do not add serum to the specimen
         Do not touch the biopsy without sterile gloves
         Do not slow-freeze. Samples should be snap-frozen (unless RNAlater® is used). Slow
          freezing promotes the formation of ice crystals, which damage the nucleic acids (e.g.,
          RNA) in the specimen. The slower a sample freezes, the larger the ice crystals. Older
          models of cryostats that require more than one minute to freeze a specimen should be
         Do not freeze a specimen directly on dry ice. If only dry ice is available, adding alcohol
          (e.g., ethanol) to the dry ice can make a slurry that will help to freeze the specimen more
          effectively (the alcohol will increase the thermal conductivity of the dry ice). This,
          however, is not a preferred snap-freezing method (see “Options for Snap-freezing”,
         Do not use dry ice alone to cool a heat extractor, if a heat extractor is used for snap-
          freezing. Dry ice has low thermal conductivity. Use liquid nitrogen instead.
         Do not attempt to take a research core biopsy if the lesion cannot be located even with
         Do not perform a biopsy if the intended biopsy site contains a hematoma from a previous
          biopsy. This is especially applicable to situations in which a second baseline biopsy for
          research is planned.
         Do not use cork and/or Whatman paper in snap-freezing, since their use may increase
          chances for contamination.

         Sterile or disposable equipment should be used, including for dissection and for snap-
          freezing or transfer of the specimen to RNAlater®.
         Instruments should be changed or cleaned between dissecting normal and tumor tissue.
         Nucleic acid amplification strategies should be used. Amplification must be applied
          across studies to avoid bias (bias is possible if some specimens undergo amplification
          and others do not).
         In the diagnostic setting, ensure that fresh/frozen research specimens do not contain
          material that would be important to the diagnosis. If they do, notify the pathologist.
         When submitting a surgical specimen to pathology, document the research biopsies that
          have been taken in the operating room. This will help to avoid misinterpretation of
          margins and other perturbations of the specimen.
         All samples intended for proteomics should be handled as consistently as possible,
          including using the same snap-freezing method.

                                             Page 29
   If a pen is used to label a cryovial or other receptacle that will be stored in freezing
    conditions, ensure that the pen is waterproof/solvent-proof and can withstand long-term
    freezing conditions.
   Copies of any relevant pathology reports and material submission forms should be sent
    along with the specimens to the central bank. Reports should be coded in a way they
    can be matched to the specimen(s) while also protecting patient confidentiality
   Any touch prep slides should also be sent to the central bank. (H&E touch prep slides
    can be shipped at room temperature.)
   Representative parts of the lesion and normal tissue should be provided. Normal breast
    tissue should be maximally distant from the tumor.
   If possible, lymph node (either negative or positive) tissue should also be provided.

                                       Page 30
Table II. Options for Snap-freezing
   Option 1: Place the specimen directly into a container of cooled isopentane. Specimen
   should freeze in 5 seconds.

   Option 2: Embed the specimen in OCT before placing it into a container of cooled
   isopentane. Specimen should freeze in 5-15 seconds.

   Option 3: Embed the specimen in OCT and place it in a plastic cryomold before placing it
   into a container of cooled isopentane. Specimen should freeze in 5-15 seconds.

   Option 4: Place the specimen in a screw-top Eppendorf-type tube, fill the tube one-third way
   with OCT, and place the tube into a container of cooled isopentane. Specimen should freeze
   in 5-15 seconds.

   Option 5: (Method used by Gentle-Jane® SnapFreezer kit. See instructions on kit for further
   detail.) Cool a heat extractor to -196°C by dipping it into liquid nitrogen. Spread a base of
   embedding medium on the specimen holder. Place the pre-cooled heat extractor onto the
   specimen holder. Specimen should freeze in 8-10 seconds.

   Option 6: Place the specimen into an empty cryovial, close the cryovial, and immediately
   submerge the cryovial into liquid nitrogen. Specimen should freeze in 30-60 seconds. (No
   OCT is used in this method.)

   Option 7: Place the specimen in a screw-top Eppendorf-type tube and place the tube
   containing the specimen into a container of cooled isopentane. Specimen should freeze
   within 30 seconds. (No OCT is used in this method.)

   Option 8: Place the specimen into a pre-labeled, pre-cooled cryovial sitting in dry ice, to
   which alcohol (e.g., ethanol) has been added (creating a slurry of the dry ice). Specimen
   should freeze within 30 seconds (max of 45 seconds). (No OCT is used in this method.)

                                         Page 31
Advantages and Disadvantages of Fresh/Frozen Tissue Collection

Fresh/frozen tissue offers a number of advantages to researchers compared to FFPE, including:

      Analysis of high-quality, intact nucleic acid (formalin fragments RNA and DNA,

       preventing efficient large-scale genome sequencing, and also results in other chemical

       modifications that present problems for biomarker research)

      Analysis of a greater number of genes (fresh/frozen tissue allows generation of more

       data from microarray-based analyses compared with FFPE)

      DNA amplification for large-scale sequencing

      Superior high-resolution array Comparative Genomic Hybridization (CGH) with DNA

      High-throughput proteomics

      Full-genome mRNA (exon analysis has not yet been adapted for FFPE RNA)

      Accurate gene expression profiling

      Protein measurements of certain antigens for which no suitable test in FFPE tissue is


      Less expensive laboratory procedures, since techniques to overcome DNA and RNA

       fragmentation are unnecessary with fresh/frozen tissue

Disadvantages of fresh/frozen tissue include the additional logistics required for sample

acquisition and preparation, particularly team-building and training, and their associated

expense, as well as safety issues associated with use of liquid nitrogen.

Examples of SOPs for fresh/frozen specimen collection and handling

Select SOPs offered for incorporation into clinical trial protocols include the following:

                                            Page 32
SOP used by BIG 00-01 / EORTC 10994 p53 study

The “Biopsy sampling and freezing procedure booklet” of the EORTC Protocol 10994

p53 study provides an easy-to-follow brochure that has been implemented successfully

across many sites of varying levels of sophistication. For adaptation in the US, some

aspects of the p53 study SOP will have to be modified for compliance with the Health

Insurance Portability and Accountability Act (HIPAA); for example, patient date of birth

and patient initials should NOT be included on the sample label.

SOP for the European Human Frozen Tumor Tissue Bank (TuBaFrost)

The TuBaFrost SOP (“Protocols for collection, storage and quality control of human

tumour        and       corresponding         normal        tissue”,     found        at


3.1.pdf) was developed in Europe for adoption by such multi-center translational studies

as those conducted by the EORTC. The SOP is meant to ensure higher comparability of

specimen quality among fresh frozen tissue samples, especially those collected from

multi-center studies. Whereas the TuBaFrost SOP emphasizes choice and flexibility with

regard to the methodology chosen for fresh frozen collection, it does offer the detailed

SOP used by the Erasmus MC Tissue Bank in the Netherlands (also provided at the

Web link above) to those sites requiring a more detailed SOP.

SOPs used for BIG 3-04 / EORTC 10041 MINDACT trial

The MINDACT trial (http://www.eortc.be/protoc/details.asp?protocol=10041) fresh/frozen

tissue collection SOPs address a range of handling procedures, broken down into easy-

to-use sets of instructions. Procedures covered by these SOPs include operating room

(OR) tumor tissue handling and tissue transportation from the OR to the pathology

department; punch biopsy sample and storage at the local pathology department; frozen

                                  Page 33
       punch biopsy sample shipment in dry ice; frozen punch biopsy sample aliquots; and

       sample registration.

       SOP used for ACOSOG- Z1031 trial

       The SOP for tissue collection from ACOSOG-Z1031, a preoperative aromatase inhibitor

       trial in breast cancer (http://www.cancer.gov/clinicaltrials/ACOSOG-Z1031), includes

       collection of both fresh/frozen and FFPE cores at baseline and at surgery (either in the

       operating room itself or in the pathology department).

Size of fresh/frozen cores

Although the larger the core the better, the minimum size for fresh/frozen tissue collection is

generally 0.25 cubic cm (approximately a 14-gauge needle). Large needles or skin biopsy tools

may be used.

It should be stressed that core biopsy – whether collected as fresh/frozen or FFPE – is superior

to fine needle aspiration (FNA) for both diagnostic and research purposes, for several reasons

(7). Foremost, cores are more representative than FNA, providing more accurate diagnostic

information – including invasive versus in situ, grade, percent tumor, and other important

histopathologic information.

Moreover, use of image guidance in core biopsy acquisition has been shown to be the superior

method for performing core biopsies, both in vivo and from excision specimens. Core biopsies

acquired under image guidance have a higher tumor yield than non-image-guided core biopsies.

This has been demonstrated by the I-SPY trial (CALGB 150007) as well as by other

experiences (7-15). Recently, it has been urged that image-guided core biopsy become the

standard diagnostic procedure for breast cancer patients who are candidates for preoperative

                                          Page 34
systemic therapy, or that this is at least the preferred method (8). Establishing and elucidating

the biology of the tumor before treatment are indeed important research objectives.

Fresh/Frozen Tissue Collection in the Diagnostic Setting

In the diagnostic setting, the collection of 2-4 fresh/frozen cores for research is recommended,

in addition to 2 cores for diagnosis.

Baseline frozen cores should be taken at the time of diagnosis. In most institutions, frozen

sections are not routine for breast cancer diagnosis. Therefore, the following alternative

approaches are recommended:

      Alternative 1: Baseline fresh/frozen cores can be taken at the time of diagnosis after the

       standard diagnostic FFPE cores have been taken. This alternative requires a second

       biopsy procedure; hence, protocol consent is obtained prior to the second biopsy.

       Specimens should be numbered as they are removed from the patient, since the first

       core often contains more tumor than subsequent cores, and frozen within 15 minutes of

       extraction. If the protocol requires the collection of both frozen and FFPE tissue, cores

       should be obtained in an alternating-core fashion (i.e., FFPE, then fresh/frozen, then

       FFPE, then fresh/frozen) as they are removed from the patient.

      Alternative 2: Baseline fresh/frozen cores can be acquired as part of the standard of care

       when the physician requests frozen-section diagnosis. These fresh/frozen cores would

       not be submitted for analysis but held in abeyance by the pathology department until the

       FFPE diagnosis has been rendered. At that point, the specimens can be released for

       research. If the routine (i.e., FFPE) samples are benign, the fresh/frozen research

       samples should be fixed in formalin, and sectioned immediately, to exclude the

       possibility that they contain important diagnostic information. On the other hand, if the

       samples indicate a diagnosis requiring surgery, then sectioning of the research biopsies

       is probably not necessary since further definitive diagnostic material will be obtained at

                                          Page 35
       surgery. The standard institutional “universal consent” for research on “excess tissue” is

       appropriate here, as long as the diagnostic work-up is considered complete when the

       specimens are released for research. (See U.S. CFR 45, Subpart 46.116, c-d for

       additional information on waivers of consent.) A subsequent, additional consent is

       administered when the patient is fully eligible for the study.

In the diagnostic setting, frozen specimens must be stored at the local site at -80°C or colder,

correctly tracked with patient identifiers, until the following have occurred: diagnosis has been

rendered on the standard sections and any needed standard biomarker panels have been

performed. Once these diagnostic issues have been resolved, specimens can be transferred to

a central facility or kept locally until they can be batch-shipped to the central laboratory.

Fresh/Frozen Tissue Collection in the Preoperative Research Setting

In the preoperative research setting, the collection of 2-4 fresh/frozen cores for research is

recommended at each of the following time points: baseline, treatment, and end of

therapy/surgery. Collection at end of therapy is especially important for studying markers of

resistance in those cancers with a significant amount of tumor remaining despite preoperative


The fraction of cores that will contain viable tumor will decrease as the tumor shrinks in

response to therapy. Therefore, fewer cores may be obtainable as lesions grow smaller.

Moreover, certain therapies put patients at risk for wound complications, which can affect the

biopsy site (16, 17).

Fresh/Frozen Tissue Collection in the Surgical Setting

During surgery, fresh/frozen tissue may be acquired in the following ways:

                                             Page 36
      Alternative 1: Transfer of the surgical specimen from the operating room to the pathology

       department in a tumor container without fixatives. Once it arrives in the pathology

       department, the sample may be sectioned, and then cores taken with a punch biopsy

       instrument and snap-frozen. This is the preferred method if ischemia time can be limited

       to 30 minutes or less.

      Alternative 2: Taking cores and immediately freezing them in the operating room – i.e., a

       core biopsy is taken from the specimen immediately after the specimen is removed from

       the patient. Operating room acquisition of cores from the specimen right after excision

       helps keep ischemia time under 10 minutes, especially in situations where pathologists

       are some distance from the operating room. Devices have been developed and are

       being piloted for frozen breast cancer specimen collection in the operating room. It is

       essential that removal of these research biopsies be documented in the submission of

       the specimen to pathology, to avoid the misinterpretation of margins or other

       perturbations of the specimen. This approach has the advantage in that pre-surgery and

       at-surgery (i.e., pre- and post-therapy) samples have been prepared in a similar manner,

       especially important in studies of preoperative systemic therapy. The ACOSOG-Z1031

       protocol provides an example scenario of acquisition of frozen cores (as well as cores

       for FFPE) before and during surgery.

The number of cores that should be taken in the surgical setting will depend on the size of the


A core biopsy approach leaves the specimen intact, thus preventing compromising of the

margin status, which can be a risk when cutting into the specimen.

If a previous core biopsy site is noted in the surgical specimen (i.e., from a core biopsy

performed before surgery), cores should not be taken from near that site. Core biopsies alter the

                                          Page 37
biology of tissue – e.g., they introduce inflammatory material from wound reaction and

biomolecules involved in wound healing – which can be problematic if a subsequent core taken

from that tissue is used for assay development. Notably, genes involved in wound healing are

very similar to those involved in cancer progression (18).

Time to Freezing and Storage Temperature for Fresh/Frozen Samples

      Fresh tissue samples, unless intended for preservation in RNAlater®, should be frozen

       as soon as possible. If they cannot be frozen immediately, they should be frozen within

       30 minutes.

      Once snap-frozen, samples should be immediately transferred either to liquid nitrogen

       (preferred) or to a -80°C freezer. If such a freezer is not immediately at hand, the

       specimen should be carried to such a freezer on dry ice.

      Frozen samples should be stored long-term either in liquid nitrogen or in a locked freezer

       with a temperature of -80°C or colder. The freezer should be alarmed, or if not, its

       temperature checked daily. If future uses of the tissue are unknown, storing the tissue in

       the vapor phase of liquid nitrogen will help to ensure long-term viability.

           o   Note: Data suggest that storage in liquid nitrogen is necessary for recovery of the

               highest quality RNA, with storage at -80°C potentially inadequate for this purpose

               (8). At the very least, liquid nitrogen should be used for storage longer than two

               years (Stephen Hewitt, personal communication, 2006).

      RNAlater® samples may be refrigerated between 2°C to 8°C for 4 days prior to shipping,

       and stored long-term at -20°C. RNAlater® samples are stable for up to one month if


      If a local site has only -20°C freezing capability, the specimen may be stored at -20°C

       until shipping, but for no longer than 1 or 2 days. RNA degradation continues to occur at

       -20C. Specimens must be stored minimally at -80C thereafter.

                                           Page 38
      Storage equipment may include small (2x3 inch) plastic, zip-top bags; mega-cassettes

       (for example, with each piece of tissue or cryomold wrapped in aluminum foil); and

       cryoboxes and plastic racks (in which cryovials can be stored).

Database Annotation for Fresh/Frozen Tissue Collection

When possible, the following should be recorded on the specimen submission form with respect

to fresh/frozen tissue collection:

      Time before freezing*

      Time before RNAlater® immersion

      Freezing temperature if not RNAlater®

      Freezing or refrigeration temperature if RNAlater®

      If previous punch biopsy(ies) were performed on the patient or noted in the specimen

      Time point of sample – e.g., after which cycle of therapy

*Note: It should be especially noted if time to freezing exceeded 30 minutes. In these cases,

time to freezing should be noted in 15-minute increments beyond 30 minutes.

Quality Assurance for Fresh/Frozen Samples

For frozen samples, a frozen section should be cut and stained with H&E to confirm tumor

presence, percent of tumor cells, preservation of morphology, and the presence of any

undesirable material, such as necrotic or inflammatory material. Some techniques involve

dividing the specimen into two pieces and taking the frozen section at the interface.

For RNAlater samples, a portion should be fixed and then stained with H&E, to assure that

tumor is present and that the tissue will be useful for IHC.

For samples intended for proteomics, a touch-prep should be performed on the cores during

sample acquisition (the I-SPY touch prep protocol is a useful reference).

                                            Page 39
Team Approach and Training for Fresh/Frozen Tissue Collection

A team approach is required for the successful implementation of fresh/frozen specimen

collection protocols. A team approach will help to ensure that site staff understands what is

acceptable practice for taking fresh/frozen cores for research.

At each site, a team should be identified and assembled that includes all staff involved in the

acquisition of these samples, including the clinician, pathologist, nurses, radiologist, surgeon,

their technical support staff, and the Clinical Research Associate (CRA). Training of the

technical support staff involved in the actual transfer and handling of fresh specimens is

essential, particularly the technical support staff for pathology, nursing, and radiology.

A backup system must be developed to account for missing personnel due to vacation/sick

leave. These personnel must be equally familiar with the protocols in place for fresh/frozen

tissue acquisition and handling.

Use of resident (physician in training) labor is problematic, since residents are seldom familiar

with these protocols, are time-constrained, not as vested in the research, over-burdened with

work, and constantly rotating. Therefore, use of dedicated tissue procurement staff with salary

support is recommended.

The Gentle Jane® instructional video provides an example of an instructional tool for

fresh/frozen tissue collection.

Additionally, kits can greatly facilitate collection of fresh/frozen samples. The Gentle-Jane®

method of snap-freezing and ACOSOG-Z1031 are examples of protocols that include kits for

fresh/frozen sample collection and freezing.

                                            Page 40
Kits should contain the necessary punch instruments, cryovials, cryomolds, RNAlater® tubes if

these are to be used, sheets, labels, color inserts, and shipping materials for collecting,

processing, and shipping the fresh/frozen specimens.


Vials, including cryovials, should be labeled with the study name or number, a specimen ID

number that is linked to the subject’s study ID, contents of the vial, and date of collection. The

latter is especially important if the sample is a serial specimen. The subject’s study ID should

not be on the vial unless patient confidentiality is determined to be secure according to the

clinical trial protocol.

Specific procedures for labeling specimens should clearly be defined in the protocol. The central

bank itself should have standardized labeling (printed or written) for archiving samples, such as

unique sample IDs and or barcodes.

For specimens that will be placed into frozen storage, information on the label should be able to

withstand -80C and not become illegible. Alcohol-based permanent markers will smudge.

Markers specifically made for cryo temperatures are recommended instead.

All levels of receptacles containing the specimen should be labeled, from the smallest unit (e.g.,

tube), to large storage units. Labeling should be resistant to cold, solvents, and water (e.g.,

cryomarker, cold-resistant label, waterproof/solvent-proof pen.)

The information included on a sample label must not include patient-identifying information, and

should be compliant with the Health Insurance Portability and Accountability Act (HIPAA). The

                                           Page 41
information should be sufficiently specific such that the encoded information (e.g., tracking

number) can be linked to the sample in the database.


In addition to the following recommendations, please also see the recommendations on

shipping contained in the “NCI Best Practices for Biospecimens Resources” (June 2007), under

Section B.1.5, “Shipping Samples”, found on page 6 of that document (1).

Shipping personnel must receive training and be certified for biological specimen shipping.

Prior to choosing a courier for frozen tissue, ensure that they 1) handle dry ice shipments, and

2) service the town/city in which the central bank is located. Dry ice should NOT be shipped with

couriers who have extremely restrictive policies concerning shipments of hazardous materials.

For international studies, each country should consider identifying a tissue bank where tissue

can be held before final shipping to a central bank across borders.

A site should consult the central bank to determine the best times to ship samples that are

frozen. This will help to avoid inadvertent thawing due to the evaporation of dry ice.

Ideally, the central bank will have included shipping materials in a kit sent to the collection site.

Batch shipping of samples will help to reduce the time required for organizing shipments and, in

the case of frozen samples, dry-ice shipping costs (see “Note on Nucleic Acid Extraction”,

below.) A good guideline for the interval of time between procurement and shipment is one


                                             Page 42
Packaging for All Specimens

Packaging should comply with International Air Transport Association (IATA) criteria (please see


If ground overnight is used for FFPE samples, then shipment should conform to ground

transportation standards (e.g., Department of Transportation packaging standards if in the US).

The shipping box should be secured and appropriate stickers should be placed, such as

“Biological Substance, Category B UN 3373”, and the type of shipment, e.g., next-day. The

IATA shipping category appropriate to the specimens collected should be used, both in labeling

and in the training required for packaging. In addition to “Biological Substance, Category B,

other IATA categories include “Exempt Human Specimens” and “Infectious Substance,

Category A”.

The latest IATA regulations should be consulted to determine which category best applies to the

specimens. Do not write the term “diagnostic specimens” on the box: revision of the IATA

regulations has replaced the term “diagnostic specimens” with “Biological substance, Category

B”. Biological substances/specimens are subject to specific packaging requirements and there

should be no misunderstanding about the contents of the shipment, particularly with regard to

risk for infection of humans or animals.

A completed material submission form from the trial needs to be submitted along with the


Packaging for Shipping FFPE Specimens

Care is required for shipping of slides and blocks. Shipment of FFPE blocks requires that the

blocks be protected from excessive heat. High outdoor temperatures are only one hazard:

                                           Page 43
placement of an unprotected paraffin block on a warm surface can result in significant damage

that could require re-embedding.

Optimally, blocks should be individually wrapped or placed in small, jewelry-size, labeled plastic

zip-top bags (not 2-10 blocks in a single sandwich bag). Blocks should then be placed in a

Styrofoam shipping container, without dry ice or cold packs. Additional space should be filled

with packing peanuts and other filler. Use of sealed bags with a desiccant can be used, if

deemed necessary, to help control humidity.

Slides should be placed in appropriate slide carriers after the Permount has dried. At a

minimum, the slide container should be wrapped in bubble wrap or placed in a padded


If alternative tissue block punchers are sent to a site for FFPE tissue collection, they should be

shipped packed into a sleeve and in secured Styrofoam.

Packaging for Shipping Frozen Specimens

Multi-level, watertight packaging with the appropriate biohazard and dry ice labels should be

used to ship frozen solid tissue and aliquoted serum or plasma. These specimens should be

contained in non-breakable – i.e., non-glass – cryovials or tube containers.

For example:

      Cryovial is placed into bubble-wrap, then into a plastic zip-top bag containing a sheet of

       absorbent material, for biohazard protection, then the bag and documentation into a

       watertight Styrofoam container packed with dry ice, the content list placed on top of

       Styrofoam container, then the Styrofoam container into a cardboard box with a

       biohazard label and a dry ice label.

                                           Page 44
Absorbent material, such as cotton balls, paper towels, or bubble wrap, should be used for

additional cushioning as needed. Fragile containers should be wrapped with cushioning

material. Again, though, plastic (not glass) vials should be used.

Shipping containers should not be sealed airtight so that CO2 gas created from the evaporation

of dry ice can escape the container. Pack dry ice and samples with paper, cardboard, or

Styrofoam so that as the dry ice sublimates the samples will not move freely inside of the

insulated box. The volume of air to which the dry ice is exposed should be minimized in order to

slow the rate of sublimation. If there is any air space after filling the package with dry ice, it

should be filled with packing peanuts or other material to reduce the volume of air space.

Reusing a dry ice box is a good use of resources. If you choose to reuse a box, completely

obliterate all unnecessary markings such as hazard labels, addresses, courier labels and


Temperature control for blood and fresh/frozen samples

      Samples should be shipped overnight, and shipped only Monday through Thursday to

       ensure delivery on a workday. If shipment cannot be made immediately, the samples

       can be stored at the appropriate temperature (e.g., -80°C for frozen tissue) until

       shipment can be made.

      Notification of shipment to the central repository is encouraged to ensure that specimens

       are properly received and processed. Communication can avoid mishaps due to

       absence or closure of the repository or variations from region to region. Tracking

       numbers and carrier information should be included in the communications.

      Dry ice should be used for shipping fresh/frozen tissue and aliquoted serum or plasma.

       The amount of dry ice needed will depend on the length of the trip and surrounding

                                           Page 45
           outside temperature, and should allow for a 24-hour delay in delivery. Discuss the

           amount required with the shipper in order to ensure that enough dry ice is added in order

           to maintain frozen specimens sufficiently to the destination.

          If dry ice is absolutely unavailable, aliquoted serum or plasma must be shipped with

           frozen gel cold packs. This should not be a routine acceptable procedure. Detailed

           documentation of the shipping condition should follow the life of the specimen.

          PAXgene samples should be sent to the central bank either at room temperature or on a

           cold pack during hot seasons, and within four days of collection.

          EDTA tubes containing blood for germline DNA extraction should be shipped on the

           same day as the blood draw (if possible), unfrozen, on a cold pack conditioned to

           maintain refrigerated temperatures during shipment. Blood should NOT be transported

           frozen, and particularly not at -20ºC.

          RNAlater® samples should be shipped on a cold pack. For more information on

           RNAlater, see http://www.ambion.com/techlib/tn/114/6.html

          The amount of refrigerant or dry ice (depending on which type of specimen is being

           shipped) should allow for a 24-hour delay in transport.

          A consideration for larger sites is the inclusion of temperature monitors within the

           shipping containers of frozen specimens to validate that temperature has been

           maintained and indicate if significant warming has taken place.

The “NCI Best Practices for Biospecimens Resources” (June 2007) provides further information

on specimen storage, under Section B.1.4, “Biospecimen Storage”, pages 4-5 of that document


                                               Page 46
Notes on Downstream Processing

Nucleic Acid Extraction

Although this guideline does not focus on actual downstream applications and processing of

tissue, the following suggestions are offered:

To ensure standardization, nucleic acid extraction should be performed at the central bank, not

at the local site, unless otherwise previously agreed to, or stated in the protocol. The

appropriate frequency and volume of such extractions have yet to be determined.

Many quality reagents and procedures are available for the isolation of nucleic acids. Therefore,

it is more important to establish a procedure with quality control measurements to determine

nucleic acid quality, than to endorse a specific manufacturer’s kit.

The quality and quantity of extracted nucleic acid should be measured and recorded, with the

metric and method for measurement also documented. Current metrics of nucleic acid quality

include OD 260/280 and Bioanalyzer analysis, and RIN number relative to ribosomal 18S and

28S. In addition to the Bioanalyzer, RNA-based assays should also include direct

measurements of quality control genes, either by RT-PCR or direct assay.

The Clinical and Laboratory Standards Institute (CLSI) guideline, “Quality Assurance for

Immunocytochemistry”, (see http://www.clsi.org and http://www.clsi.org/source/orders/free/MM4-

a.pdf) provides consensus recommendations and an extensive amount of information on

immunocytochemical assay design, validation, and performance, as well as on tissue handling

and quality (19). This document, from 1999, is currently under revision, with additional guidance

expected in CLSI document IL-28a.

                                            Page 47
Whole blood should be shipped fresh (unfrozen) on a cold pack, via priority overnight service

(i.e., within 24 hours at 4°C) to the central bank for DNA extraction. The amount of DNA isolated

from blood samples can be highly variable. DNA yield depends on the number of white blood

cells present and blood storage conditions. On average, 10 mL of fresh blood should yield

approximately 350 g of DNA (Mark Bouzyk, personal communication, August 17, 2007).

Once at the central bank, germline DNA should be extracted from the blood within five days of

the blood draw, ideally within 24 hours, and this DNA should be stored between -20ºC and -


If blood must be frozen before DNA extraction, the blood should be stored at -80°C to ensure an

acceptable DNA yield. Since this guideline recommends that all of the recommended whole

blood collection be used for DNA extraction, aliquoting of whole blood should not be necessary.

In situations where not all blood is intended for extraction, however, aliquoting into plastic

centrifuge tubes and freezing at -80°C would be appropriate, to avoid freeze-thaw. It should be

noted that, whereas blood samples frozen at -80°C typically exhibit some decrease in DNA

yield, samples frozen at -20°C exhibit a further decrease; thus, storage at -80°C rather than -

20°C is recommended (Mark Bouzyk, personal communication, August 17, 2007).

For FFPE samples, batch extractions using the same isolation protocol are preferable, to help

ensure consistency of methodology. Even if isolation procedures have improved over time, the

same isolation procedures should be used for each batch extraction of specimens collected

from a given clinical trial. Biomarker analyses may best be carried out on freshly cut FFPE

sections, due to antigen loss in stored FFPE sections (2).

The method used to deparaffinize FFPE tissue has an important impact on RNA recovery.

Chung et al. noted improved (by 2-fold) recovery of RNA through use of higher temperatures

                                          Page 48
and longer (3-day) lysis times when deparaffinizing blocks (6). Current extraction protocols may

suffer from inadequate deparaffinization (20).

For fresh/frozen tissue, core samples should have histological evaluation and disease

confirmation (e.g., for presence of tumor, any necrotic or inflammatory material, and other tissue

features) prior to extracting biomolecules. The SOP described by Schmitt et al. is a useful

reference for the disintegration of fresh frozen tissue for biomarker extraction and assessment

(21). Upfront isolation and storage of RNA from fresh/frozen tissue has been advocated, as

opposed to later extraction, since RNA has been shown to degrade within stored tissue (6).

Once extracted, RNA should be stored at -80°C or colder (e.g., liquid nitrogen), while DNA can

be stored between -20ºC and -80ºC. RNA/DNA should be stored in freezers that are either

alarmed or checked daily for temperature maintenance. Jänicke et al provide an example of a

multi-center clinical trial whose treatment arm assignments were based on results from

biomarker testing on fresh/frozen samples (22).

Note on Proteomics

It remains to be determined what methodology to recommend for the handling and processing

of fresh/frozen tissue for proteomics. The EORTC PathoBiology Group SOP described by

Schmitt et al. for assessment of protein markers such as uPA/PAI-1 from fresh/frozen tissue,

however, provides useful guidance (23).

Likewise, guidelines on proper collection and processing of blood for proteomics are needed.

This area of research is still a substantial unknown. Mitchell et al. and Rai et al. are useful

references regarding this area of investigation (5, 24).

                                            Page 49
Quality Assurance of FFPE Tissue by the Central Bank

All FFPE blocks submitted for research should be subjected to quality assessment and quality

control measures. This QA/QC is important not only for research, but also to protect patients in

cases where research specimens contain important diagnostic information.

At the central specimen bank for a clinical trial, the following should be performed on an FFPE


        A top section should be cut for H&E review by a pathologist. H&E review should confirm

         the following:

         o   Presence, percent, and surface area of a) tumor, b) normal, and c) in situ tissue in

             the block (what constitutes an “acceptable” percent of tumor will be study-


         o   Size of tissue, expressed in square mm or total tissue size (noting size is necessary

             since percent alone does not convey amount, and a single tissue block tumor size is

             not the same as pathology report grossing tumor size).

         o   Morphology of the cancer has been confirmed and that comments on quality have

             been provided by the reviewing pathologist.

         o   Diagnosis, including grade and lymphatic invasion

         o   “Interfering” material, such as fibrosis, necrosis, inflammation, mucin, and poor

             fixation (in the absence of normal DNA, however, infiltrating lymphocytes may

             actually be a valuable source of normal DNA).

         o   Areas of interest consistent for TMA coring, including normal tissue, should be

             marked by a pathologist, using color code.

        If many sections are cut, H&E staining may have to be performed at intervals to ensure

         representation of tumor. For many IHC assays, this can be done on the IHC section.

         How frequently H&E sections are taken will depend on the study. One suggestion is to

                                            Page 50
       take an H&E from an individual tissue block after every 10 to 12 five-micron sections,

       e.g., a sequence of 1 H&E and 10 unstained slides (55 microns).

      Guidance on quality assurance for TMAs is provided in this Appendix (“Tissue

       Microarray (TMA) Construction and Quality Assurance”).

Conservation of FFPE tissue can be furthered in the following ways:

      At least two representative unstained sections and an H&E slide from each used block

       should be stored in a permanent file to ensure that tissue is available in case of patient

       medico-legal need.

      Construction of tissue microarrays (TMAs) will help to conserve tissue (see “Tissue

       Microarray (TMA) Construction and Quality Assurance”, below).

      Taking and storing cores for DNA/RNA extraction during TMA construction (noting the

       tumor and normal portions of these cores will be important if they are used, for example,

       in developing tumor-only assays).

      An experienced histotechnologist should be assigned to all aspects of handling blocks.

      Sectioning:

       o   Blocks should be sectioned in a consistent manner by an experienced

           histotechnologist, and a second experienced technologist assigned to assess the

           quality of the cut sections. Sectioning by less experienced technicians can result in

           many microns of block lost to re-facing.

       o   Tissue sections should not be greater than 3 to 5 m.

       o   Disposable microtome blades should be used to prevent cross-contamination.

       o   Use of a block alignment instrument such as a Histo-Collimator can reduce loss of

           tissue when re-facing the block during the cutting process.

       o   Re-facing trims should be collected into microfuge tubes for future nucleic extraction

           studies where LCM is not required.

                                           Page 51
      Specimens should only be released to investigators after review of the scientific merits of

       the proposed use. Review by independent scientific advisors is recommended.

If excess water is noted on tissue sections, the water should be removed, and the sections

oven-baked for 15-30 minutes at 65°C before the sections are used for an assay.

Tissue Microarray (TMA) Construction and Quality Assurance by the Central Bank

The following is based largely on the “PCO Best Practices and Operations Roadmap” 2005 draft

document developed by the National Cancer Institute Cooperative Group Banking Committee

and the Cancer and Leukemia Group B.

TMAs are the key sources of material for retrospective studies that serve as the first steps in the

discovery and validation of new prognostic and predictive biomarkers. Coring for TMAs is

especially important for those blocks that must be sent back to the collection site.

Construction of TMAs from large adjuvant trials is strongly encouraged. Prospective TMA

creation on all studies, both for banking and study-specific design, are strongly recommended

prior to whole sectioning of blocks for other non-TMA projects.

Advantage of TMAs

The major advantage of TMAs is that they allow the performance of tissue-based assays

(immunohistochemistry, histochemistry, in situ hybridization, etc.) on a large number of patient

samples in an efficient and cost-effective manner.

                                           Page 52
With TMA technology, several hundred representative cores from several hundred patients may

be included on a single glass slide for assay. Thus, significantly more tissue can be conserved

than if the block were sectioned serially.

Disadvantage of TMAs

The major disadvantage of TMAs is that each core (or set of cores) represents a fraction of the

lesion. It should be noted that TMAs are not suitable for all lab studies – e.g., microvessel

density staining and measurement.

Pathologists must be involved in protocol development early on, so that questions regarding

TMA design can be addressed during the protocol development. This will help to ensure that

there are no special constraints or issues related to the type of tumor in question.

Documentation of decisions made for the TMA design should be kept by the trial’s central bank.

Pathologists from the trial’s leadership should review requests for TMA construction. These

should include pathologists from the Group’s Pathology Coordinating Office (PCO) and the

Group pathology committee.

Technical guidance on TMA construction

The following are recommended for breast TMA construction:

      Before constructing the TMA, make a TMA map. This map will also serve as a guide for

       future sectioning requests.

      The area of the block to be cored for a TMA should be selected by a pathologist. A fresh

       H&E section (3-5 µm) from the block should be used for this determination. Areas to be

       sampled (tumor, normal, and pre-malignant tissues) should be identified. The

       appropriate sampling site should be circled or otherwise marked on the corresponding

       H&E slide or directly on the block before the array process is begun.

                                             Page 53
   Unless otherwise indicated by the protocol, tissue core samples for the TMA should be

    removed from the submitted FFPE blocks prior to the sectioning or other tissue

    processing of a tissue block, for example for correlative science or nucleic acid

    extraction (if too many sections are cut before TMA construction, there is a dramatic loss

    of targets). Sections for clinical purposes are excluded from this requirement.

   Following selection by a pathologist, random sampling of the region of interest from the

    same block should be used for the core selection. Random sampling is especially

    important for breast cancer. (Note: Random coring may not be appropriate for other

    types of cancer.)

   TMA core size and number of cores should be determined by the pathology committee

    for the trial, for example by the Pathology Committee of the Cooperative Group leading

    the trial. Use of smaller core diameters, however, allows for a greater number of cores to

    be extracted from the lesion and a greater number of cores that can fit into the TMA

    block. In general, taking three cores (1.0 mm2 size preferred; 0.6 mm2 is adequate) is

    recommended, with placement on at least triplicate TMAs (see “Placement of same-

    patient cores in same or separate blocks”, below, for more information). Greater tumor

    representation and increased ease of handling is expected as core size increases.

   No fewer than three tumor cores should be selected from each tissue block, if size

    permits. The optimal numbers of cores needed to address the issue of intratumoral

    variability will be a function of the marker of interest and the tumor type. In general, the

    greater the degree of intratumoral heterogeneity for any given marker, the greater the

    number of cores required (25).

   Areas of necrosis, inflammation, or fibrosis should be avoided unless required by the


   When coring for TMA, an extra core should be taken for nucleic acid extraction (note that

    this may not qualify for tumor-only downstream assays).

                                        Page 54
   Note: Unless microdissection is used, cores may be as good as, if not better, than

    sections for extraction of RNA and DNA.

   After each donor block is punched, cut 1 section and stain with H&E, and file this


   Replicate TMA blocks should be created (three, four or more replicates). Ensure that

    replicates are made such that triplicate or multiple cores are placed on separate blocks –

    this allows selection of the appropriate number of cores per marker.

   The maximum number of cores that should be placed on a single block will vary

    depending on core size, block size, and IHC methodology, among other factors. You will

    want to avoid placing so many cores on a TMA that the surface section of cores

    becomes larger than the antibody coverage area on the slide programmed by the

    autostainer, or that the cores are on the periphery of the slide and difficult to observe or


   Density: The distance between cores should NOT exceed the core diameter (otherwise,

    manual interpretation is hindered).

   Controls should be placed on each TMA block, for quality control and to address tumor


    The following may be used as controls:

       o     Tissue-specific controls: i.e., breast cancer cell lines. Typically, the more cell

             lines used, the better. Three to four cell-lines in duplicate are recommended.

             Inclusion of SKBR3, MCF7, T47D, and breast cancer cell lines from the NCI 60 is


       o     Biology-associated controls: e.g., salivary gland, endometrial (hormone marker),

             testis (proliferation marker), fibroadenoma.

       o     Organ system controls: adrenal gland, brain, breast, colon, kidney, liver, lung,

             pancreas, placenta, prostate, testis, uterine myometrium (smooth muscle).

                                          Page 55
                  Note: Normal tissue, at a minimum, should contain: liver, kidney,

                  endometrium, lymph node, colon, and testis (note that normal breast tissue is

                  not included here).

   Note: Although controls do not account for staining variability from different inter-institutional

   fixation practices, if an assay produces consistent results despite a variety of samples, this

   attests to its robustness.

      Quality assurance measures for the TMA should include the following:

          o   Perform H&E on every fiftieth (50th) slide taken from the TMA blocks.

          o   Perform ER, PR and HER2 on the TMA tissue upon sectioning and before

              distribution of TMA sections to researchers, even if data are available from the

              original pathology.

              Central ER/PR/HER2 assessment will be necessary until such a time that ER,

              PR and HER2 are performed in a standardized fashion across institutions.

              Central assessment as well as data collection MUST be done via the central

              banks/labs since future studies will compare findings with ER, PR, and HER2

              status of the tissue in order for the study to be meaningful.

          o   The above tests should be reviewed by a pathologist.

Placement of same-patient cores in same or separate blocks:

The decision to place cores from the same case either in the same TMA block or distributed

among multiple TMA blocks, or a combination of both, depends on a variety of factors.

“Pro’s” for placement of cores from the same case onto different TMA blocks:

      Addresses the problem of edge effect.
                                           Page 56
      If a particular block becomes damaged or exhausted, patient material from the other

       blocks would still be preserved.

      If a marker is fairly heterogeneous, then one would want to look at a given case’s cores

       across multiple blocks. For markers that are not very heterogeneous, one would first test

       with one block, then validate with the other block(s).

      Likewise, placing same-patient cores on different blocks can reduce bias in interpretation

       of intratumoral variability – for example, if the same type of staining persists across the

       same patient’s cores placed in different blocks; this increases the validity of the


“Pro’s for placement onto the same block:

      Morphology is a continuum, and tumor heterogeneity can be extreme. One would not

       want to wonder if, for every core, the cores from the same case that were placed on the

       other blocks should also be examined.

      If tumors are small and heterogeneous, same-block placement may be preferable. (For

       non-breast tumors such as gastrointestinal tumors in particular, one would want to look

       at middle, invasion front, etc.)

In a well-constructed TMA, core dropout should be the same regardless of which of the above

two methods is chosen (Stephen Hewitt, personal communication, 2006).

TMA layout

TMA layout should be asymmetric and irregular so that it is relatively easy to orient the TMA

block. For example, the following features may be included:

      Blank rows and columns that do not run down the center lines of the TMA blocks

                                            Page 57
      A blank corner

      Asymmetric distribution of control cell lines and tissue controls

      Labeling of one side of the TMA block

Cores should start 3 mm away from the block edges, to prevent the paraffin from cracking.

If same-patient cores are to be placed on the same block, they should be dispersed on the

block. This will decrease the risk of interpretation bias. Same-patient cores should not be placed

adjacent to each other, since this can bias interpretation.

If same-patient cores are to be dispersed across multiple blocks, it may be desirable to place

them in different regions of the array (outer and inter), with random placement, rather than

placing them in the same location in each block. It must be noted that mapping and multi-plate

recipient setup may not permit random placement.

Whereas “random designs” can help to prevent interpretation bias, they can create other

problems for pathologist readers and data integration with automated image analysis.

Mini arrays of the cores (3x5, 4x5, 5x5, and 6x5) can be spaced for easy orientation, with

control tissue in the rows between the mini-arrays.

Placing stained cores of “irrelevant” tissue at the edge of the grid can be useful to mark

orientation: for example, red at one corner, and green at opposite end of array.

                                            Page 58

1. National Cancer Institute Best Practices for Biospecimen Resources
    http://biospecimens.cancer.gov, National Cancer Institute Office of Biorepositories and
    Biospecimen Research, June 2007
2. DiVito KA, Charette LA, Rimm DL, et al: Long-term preservation of antigenicity on tissue
    microarrays. Lab Invest 84:1071-8, 2004
3. Hewitt, et.al., Quality Issues in the Recovery of RNA and DNA from Archival Tissue. Tissue
    Handling and Specimen Preparation in Surgical Pathology: Issues Concerning the Recovery
    of Nucleic Acids from Formalin Fixed, Paraffin Embedded Tissue: A Report from Paraffin
    Embedded Working Group. Submitted
4. Fergenbaum JH, Garcia-Closas M, Hewitt SM, et al: Loss of antigenicity in stored sections
    of breast cancer tissue microarrays. Cancer Epidemiol Biomarkers Prev 13:667-72, 2004
5. Mitchell BL, Yasuie Y, Lia CI, et al: Impact of Freeze-thaw Cycles and Storage Time on
    Plasma Samples Used in Mass Spectrometry Based Biomarker Discovery Projects. Cancer
    Informatics 1:98-104, 2005
6. Chung JY, Braunschweig T, Hewitt SM: Optimization of recovery of RNA from formalin-fixed,
    paraffin-embedded tissue. Diagn Mol Pathol 15:229-36, 2006
7. Singh B: "Initial pathology assessment prior to preoperative therapy". Presented at the
    National Cancer Institute Meeting: "Preoperative Therapy in Invasive Breast Cancer:
    Reviewing the State of the Science and Exploring New Research Directions", Natcher
    Conference Center, Bethesda, Maryland.
    http://ctep.cancer.gov/bcmeeting/index.html#agenda (See also "Wrap-up 'statement of the
    science'" presented by Gralow J, Winer E. at the same meeting). March 26-27, 2007
8. Fajardo LL, Pisano ED, Caudry DJ, et al: Stereotactic and sonographic large-core biopsy of
    nonpalpable breast lesions: results of the Radiologic Diagnostic Oncology Group V study.
    Acad Radiol 11:293-308, 2004
9. Fuhrman GM, Cederbom GJ, Bolton JS, et al: Image-guided core-needle breast biopsy is an
    accurate technique to evaluate patients with nonpalpable imaging abnormalities. Ann Surg
    227:932-9, 1998
10. Liberman L: Percutaneous image-guided core breast biopsy. Radiol Clin North Am 40:483-
    500, vi, 2002
11. Miller JC: Percutaneous Image-Guided Breast Biopsy. Radiology Rounds: A Newsletter for
    Referring Physicians Massachusetts General Hospital Department of Radiology, 2006 pp 1-
    4 (Newsletter)
12. Perlet C, Heywang-Kobrunner SH, Heinig A, et al: Magnetic resonance-guided, vacuum-
    assisted breast biopsy: results from a European multicenter study of 538 lesions. Cancer
    106:982-90, 2006
13. Rosen M, Marquez H, Herman B, et al: Factors affecting quality of tumor core biopsy
    specimens in the CALBG/ACRIN neoadjuvant breast (I-SPY) trial. Abstract 6006, San
    Antonio Breast Cancer Symposium (SABCS). San Antonio, Texas. , 2006 (Abstract)
14. Simon JR, Kalbhen CL, Cooper RA, et al: Accuracy and complication rates of US-guided
    vacuum-assisted core breast biopsy: initial results. Radiology 215:694-7, 2000
15. Verkooijen HM: Diagnostic accuracy of stereotactic large-core needle biopsy for
    nonpalpable breast disease: results of a multicenter prospective study with 95% surgical
    confirmation. Int J Cancer 99:853-9, 2002
16. Tan AR, Yang X, Hewitt SM, et al: Evaluation of biologic end points and pharmacokinetics in
    patients with metastatic breast cancer after treatment with erlotinib, an epidermal growth
    factor receptor tyrosine kinase inhibitor. J Clin Oncol 22:3080-90, 2004
17. Wedam SB, Low JA, Yang SX, et al: Antiangiogenic and antitumor effects of bevacizumab
    in patients with inflammatory and locally advanced breast cancer. J Clin Oncol 24:769-77,

                                         Page 59
18. Riss J, Khanna C, Koo S, et al: Cancers as wounds that do not heal: differences and
    similarities between renal regeneration/repair and renal cell carcinoma. Cancer Res
    66:7216-24, 2006
19. O'Leary TJ, Edmonds P, Floyd A.D., et al: Quality Assurance for Immunocytochemistry;
    Approved Guideline. NCCLS (now Clinical and Laboratory Standards Institute) document
    MM4-AC (ISBN 1-56238-396-5). http://www.clsi.org/source/orders/free/MM4-a.pdf (See also
    http://www.clsi.org), December 1999
20. Faolain EO, Hunter MB, Byrne JM, et al: Raman spectroscopic evaluation of efficacy of
    current paraffin wax section dewaxing agents. J Histochem Cytochem 53:121-9, 2005
21. Schmitt M, Mengele K, Schueren E, et al: European Organisation for Research and
    Treatment of Cancer (EORTC) Pathobiology Group standard operating procedure for the
    preparation of human tumour tissue extracts suited for the quantitative analysis of tissue-
    associated biomarkers. Eur J Cancer 43:835-44, 2007
22. Janicke F, Prechtl A, Thomssen C, et al: Randomized adjuvant chemotherapy trial in high-
    risk, lymph node-negative breast cancer patients identified by urokinase-type plasminogen
    activator and plasminogen activator inhibitor type 1. J Natl Cancer Inst 93:913-20, 2001
23. Schmitt M, Sturmheit AS, Welk A, et al: Procedures for the quantitative protein determination
    of urokinase and its inhibitor, PAI-1, in human breast cancer tissue extracts by ELISA.
    Methods Mol Med 120:245-65, 2006
24. Rai AJ, Gelfand CA, Haywood BC, et al: HUPO Plasma Proteome Project specimen
    collection and handling: towards the standardization of parameters for plasma proteome
    samples. Proteomics 5:3262-77, 2005
25. Kim WH, Rubin MA, Dunn RL: High-density tissue microarray. Am J Surg Pathol 26:1236-8;
    discusson 1237-8, 2002

                                          Page 60

To top