Setting up a Prospective Thyroid Biobank for Translational Research: Practical Approach of a Single Institution (2004

Abstract

In the last few years, conditions for setting up a human biobank in France have been upgraded by taking into account (1) the new laws and regulations that integrate the ethical and societal dimension of biobanking and delineate the risks for patients associated with the procurement of human cells and tissues, (2) the increasing request by scientists for human samples with proven biological quality and sophisticated sets of annotations, including information produced through the evergrowing use of molecular biology in pathology, and (3) establishment of procedures concerning the safety of the personnel working with biological products. For this purpose, health authorities and national research institutes in France have provided significant support for the set up of biobanks. The present work was conducted to describe how we set up a biobank targeting diseases of a specific organ (thyroid gland), with the aim of rapidly developing translational research projects. The prospective experience of a single institution (Pasteur Hospital, Nice, France) over a 6-year period (2004–2009) is presented from the practical point of view of a surgical pathology laboratory. We describe different procedures required to obtain high-quality thyroid biological resources and clinical annotations. The procedures were established for the management of biological products obtained from 1454 patients who underwent thyroid surgery. The preanalytical steps leading to the storage of frozen specimens were carried out in parallel with diagnostic procedures. As the number of international networks for research programs using biological products is steadily increasing, it is crucial to harmonize the procedures used by biobanks. In this regard, the described thyroid biobank has been set up using criteria established by the French National Cancer Institute (Institut National du Cancer) to guarantee the quality of different collections stored in biobanks.

Introduction

Our current knowledge of mechanisms of human diseases has been greatly improved by the onset of new molecular analytical approaches performed with cells, tissues, and other biological materials. It is still better to work on frozen samples rather than on fixed samples when using biotechnological methods and approaches such as DNA, miRNA, and cGH arrays, the real-time polymerase chain reaction (PCR), and proteomics.^1–3 Alternatively, fresh tissue from resection specimens can be stored in RNA Later (Ambion, Austin, TX) for a few hours during preliminary dissection and then frozen in liquid nitrogen.⁴ In this regard, it is crucial to set up a highly organized structure to perform the sample-freezing procedures, to control the preanalytical steps, and to efficiently store multiple types of specimens in biobanks.⁵

As the final goal of translational research is not only to identify new cellular and molecular mechanisms that lead to the onset, or modify the course of a disease, but also to improve treatment and establish preventive programs, the setting up of such biobanks is of key importance.⁵ In this regard, substantial efforts have been made in recent years to build a network of human biobanks in France. A significant number of projects were supported by different French institutions to set up biological collections for clinical and translational research (www.e-cancer.fr/Espace-tumorotheques; www.crbfrance.fr). Many of these human biological collections mostly include frozen tissue specimens.⁶ Thus, the criteria for collection and storage of human biological products of high quality have been established.⁷ As some of these criteria were recently redefined, a number of collections that were set up a few years ago do not currently fulfill all new defined criteria.⁷ Thus, numerous frozen human tissues and other biological materials collected by many biobanks in France cannot be fully exploited, particularly for international network projects and/or for current programs that use sophisticated biotechnological approaches, such as deep sequencing technology.⁸ Moreover, a great number of frozen specimens stored in these biobanks have probably not been collected with the signed consent of patients and thus cannot be used for clinical research in Europe.

Despite the large number of human biobanks that exist all around the world and the importance of biobanking for the development of translational research, there are only a few reports reviewing the practical aspects of setting up such a biobank. Since 2004 we set up a prospective human biobank, focusing on a limited number of diseases and organs, particularly the thyroid gland (http://www.biobank06.com).⁹ For this purpose, we used the different criteria defined and recommended by the French National Cancer Institute [Institut National du Cancer (INCa)],⁷ to which we added complementary clinical and pathological criteria for this organ. Moreover, we collected frozen tissue, plasma, serum, constitutional DNA, and embedded paraffin tissue for tissue microarrays (TMAs) from each patient and developed primary cultures of thyrocytes from certain specimens of interest.

This article is intended to provide our view of “best practices” referring to thyroid pathology biobanking. We address issues related to the collection and annotation of thyroid samples and the ethical and administrative aspects of biorepository management. Through the evaluation of our experience (Pasteur Hospital, Nice, France) from 2004 to 2009, we provide clues on how to prospectively set up an organ-oriented biobank that can be rapidly used for translational research as a high-quality resource for both nontumoral and tumoral diseases.

Patients and Methods

Ethics, consent, informative procedures, and confidentiality

This study was conducted between September 1, 2004, and December 31, 2009, in the biobank of the Pasteur Hospital in Nice. This biobank is certified according to NF 96S-900 certification (No. 2010/36548) and Inserm (Institut National de la Santé et de la Recherche Médicale) and IBiSA (Infrastructures en Biologie Sante et Agronomie; www.ibisa.net) criteria. Moreover, all patients hospitalized in the Department of Oto-Rhino-Laryngology (Pasteur Hospital) and who underwent thyroid surgery were systematically included in this study. Patients were informed by the surgeons (N.G., L.C., J.S.) of the reason for the different procedures and the goals in setting up a biobank. An information leaflet was provided to patients before surgery. A consent form was signed by both the patient and the surgeon before surgery, establishing permission to obtain frozen thyroid specimens, plasma, sera, constitutional DNA from blood, primary thyrocytes for culture, and embedded paraffin tissue for TMA construction.¹⁰ The patients were informed that all data, particularly clinical data, stored in the biobank were anonymized to protect patients. The different procedures concerning the anonymization of patients' data were approved by the CNIL (Commission Nationale de l'Informatique et des Libertés; www.cnil.fr).

The thyroid specimens and derived products from blood patients were then systematically frozen in the surgical pathology laboratory [Laboratory of Clinical and Experimental Pathology (LCEP), Nice, France] and stored in the biobank unit.

Clinical and epidemiological data and patient follow-up

The clinical and epidemiological data were obtained by the physician during the preoperative visit of the patient. Collected items include the family and patient's medical history, physical symptoms, thyroid ultrasound results, main biological tests, and the results of thyroid fine-needle biopsy aspiration (an example of the recorded items can be seen at http://www.biobank06.com). After histological diagnosis, the therapeutic strategy was recorded. Patient outcome was systematically evaluated and updated every 6 months.

Tissue specimen management

After resection, each surgical thyroid specimen was placed in a plastic sterile container (125 mL; Dutscher, Brumath, France) and then in an envelope for vacuum transport (Décomatic, La Verpillière, France) from the Department of Oto-Rhino-Laryngology to the LCEP through a pneumatic tube. The time elapsed between the surgical procedure and freezing was recorded. Surgical thyroid specimens were photographed before freezing and these images were recorded in a database.

Specimens indicated for freezing procedures were obtained during the preparation of frozen sections for diagnosis, all else was routinely frozen. Selection of the different specimens was performed by a senior surgical pathologist (S.L., V.H., M.I., C.B., P.H.). This was done under a hood, with sterile gloves and a sterile razor blade, on a dedicated surface area cleaned with RNAse up solution (Molecular BioProducts, San Diego, CA). All scissors and forceps were cleaned with RNAse up solution before use. Specimens taken for freezing were selected by macroscopic examination of gross lesion(s) and normal tissue. In the case of several lesions, they were numbered (lesion 1, lesion 2, etc). Depending on the size of the lesion, between 1 and 10 specimens were frozen in liquid nitrogen (LN₂). The size of frozen specimens depended on the size of the lesion(s). However, most of the frozen specimens had a volume of around 0.5 cm³. In parallel, and whenever possible, thyroid specimens (from 1 to 10) from macroscopically nonaffected areas were similarly processed. Finally, all specimens were stored in cryotubes (1 specimen per tube) (1.8 mL, Nunc cryotube; Dutscher) and weighed immediately before freezing. Among the different frozen specimens in LN₂, 1 was used for subsequent RNA extraction (first specimen to be frozen) and 1 was used for subsequent DNA extraction (last specimen to be frozen). When regional lymph node excision was performed, specimens were similarly frozen for biobanking after frozen sections were obtained for diagnosis. The different lesions and areas to be frozen in LN₂ were schematically reproduced on a sheet to give a thyroid cartoon. During the freezing procedure, the mirror image of the frozen tissue specimen was taken for formalin fixation. A deparaffinized section of this tissue mirror embedded in a paraffin block was stained with hematoxylin and eosin. Then, the mirror images of representative samples of snap-frozen thyroid lesions and normal adjacent tissue were photographed, and all images were archived with the biobank software (Cresalys Software; Alphelys, Paris, France). The samples obtained from patients with a known associated infectious pathology (human immunodeficiency infection, hepatitis B and C, tuberculosis, etc.) were stored separately from samples obtained from other patients. Some frozen specimens with potential value for improving diagnosis and/or prognosis (particularly for lymphoma and sarcoma) were separately managed. In this regard, registration with biobank software of these latter frozen specimens was linked to supplementary information saying that these rare specimens can be used immediately to obtain data for improvement of diagnosis and/or prognosis (for example, by doing molecular biology analyses) if necessary. The use of these specimens for research projects requires more than 1 frozen specimen. Finally, the main data were recorded and summarized on a single datasheet (www.biobank06.com) (Fig. 1). The different criteria allowing inclusion of thyroid tissue in the database are shown in Table 1.

FIG. 1.

Example of a datasheet for recording the main data for inclusion. A color version of this figure is available in the online article at www.liebertonline.com.

Table 1.

Criteria for Inclusion of a Thyroid Specimen into the Database

“Mandatory” criteria

1. Obtain signed consent from the patient

2. Time between resection and freezing of tissue less than 30 min

3. All specimens weighed in cryotubes before freezing

4. Specimens taken from both lesion and adjacent nonlesional areas

5. Digitization of images from mirrored samples fixed in formalin

6. Core of embedded paraffin tissue embedded in tissue microarrays

7. Complete clinical data available

8. Complete histological data available

Highly recommended criteria

1. Plasma and sera samples available

2. At least 7 frozen lesional and adjacent lesional specimens

3. Specimens dedicated to RNA and DNA extraction

4. Genetic profiling available for tumor pathologies

5. Primary cultures available for corresponding frozen specimens

Pathological data

Gross macroscopic and histological data and pathological tumor node metastasis staging were recorded on a sheet that was filled in by pathologists. The list of these items is available at www.biobank06.com. A subset of this collection appears at www.paca-biobank.com.

Tissue microarrays

The tissue specimens included in paraffin blocks were used to prepare TMAs that are specific for different thyroid pathologies. Depending on the size of the initial lesion, 0.6-mm spots were obtained in triplicate using an arrayer (Alphelys). At least 4 identical TMA paraffin blocks were set up in parallel using a booster system (Alphelys).

Blood sample management

Since October 2006, blood samples were collected in parallel in the biobank. A sample of blood was obtained from patients before surgery. Two tubes, 1 with ethylenediaminetetraacetic acid containing some tripotassium ethylenediaminetetraacetic acid 3 (Greiner Bio-One, Frickenhausen, Germany) and another without anticoagulant, with 10 mL of blood were transferred to the LCEP through a pneumatic tube. Blood was immediately centrifuged, and plasma and sera were collected under a hood in a culture room.

Primary culture of thyrocytes

A collection of primary cultures of thyrocytes was set up from certain surgical specimens representing diverse diseases (conventional papillary thyroid carcinomas, follicular variant of papillary thyroid carcinomas, tumors of uncertain malignant potential, adenomas, Graves' disease, and normal thyroid tissue). For this purpose, thyroid tissue was obtained from 39 human subjects undergoing surgery for solitary or multiple nodules. Nodules were removed and separated from healthy tissues. Both affected and unaffected tissues were used for the preparation of cells for culture in vitro. Follicles were prepared and plated on Petri dishes as previously described.¹¹

Quality assessment and nucleic acid collection

RNA and DNA collection were set up in parallel. One specimen of affected thyroid tissue and 1 specimen of presumably healthy thyroid tissue were specifically dedicated to the nucleic acid biobank. RNA was extracted using a Trizol procedure. The RNA quality was evaluated from the 28s/18s ratio and the RNA integrity number (RIN) on an Agilent Bioanalyzer 2100 (Agilent Technologies, Santa Clara, CA). Similarly, DNA was extracted from another pair of tissue samples. The DNA quality was measured by PCR amplification of a fragment of 950 bp. Extraction was done from 1 selected specimen of each tumor from 2004 to 2006 to set up stocks of nucleic acids. However, because of time and cost limitations, since January 2007, nucleic acid extractions were then performed once a week from 1 specimen only, to control the quality of the freezing procedure (particularly the time of cold ischemia). Moreover, once a month, 1 frozen specimen stored at −80°C for 6 months was taken, and both the DNA and RNA quality was evaluated by either PCR amplification or evaluation of the RIN, respectively, and recorded.

Storage facilities and safety procedures

Cryotubes containing tissue specimens were initially placed in boxes (cryoboxes, which can contain 81 cryotubes; Dutscher) and stored in an LN₂ container with a temperature recording system (Taylor-Warthon, Husum, Germany). However, a “backup” empty LN₂ storage tank was available in the facility. Once a week, the cryoboxes were placed in −80°C freezers dedicated to thyroid tissue collection. Freezers were equipped with a permanent temperature recording system and an independent centralized alarm. Two backup empty −80°C freezers were available and a technician was on call in case of failure. Plasma and sera were stored in a −140°C freezer with an identical security system as that for the −80°C freezers. One backup empty −140°C freezer was available. RNA and DNA samples were stored in a −80°C freezer. Finally, primary cultures were stored in a nitrogen tank. The alarm was set to detect temperature increases above −65°C for the −80°C freezers and −130°C for the −140°C freezers. General access to the biobank unit and to the LCEP is regulated by personalized magnetic cards. Access to the cryoconservation room is strictly controlled and limited to staff members with a digicode system for regulating access. The opening of the different freezers was regulated and recorded. Paraffin blocks and slides were stored at room temperature and TMAs at 4°C. For these latter samples, the refrigerator had permanent temperature recording and an alarm setup to detect temperature increases above 8°C. Between 6 pm and 8 am, the biobank unit is protected by a central alarm and digicode access. The general organization of the surgical pathology laboratory (LCEP) and biobank unit space can be seen at www.biobank06.com and is shown in Fig. 2.

FIG. 2.

Floor plan of the biobank unit and the surgical pathology laboratory (Laboratory of Clinical and Experimental Pathology). A color version of this figure is available in the online article at www.liebertonline.com.

Information technology

The information technology system allows patient registration, specimen tracking, biological material cataloging, quality control, and specimen availability. Thus, the clinical, epidemiological, pathological (including images), and biological information and the inventory of tissues and biological products are managed with the same specific software (Cresalys Software; Alphelys). An example of available information can be accessed at www.biobank06.com (CHU of Nice). A summarized datasheet can be accessed for each patient (Fig. 1) (www.biobank06.com). Supplementary information can be accessed at www.biobank-paca.com. However, all available information for each patient is stored using the biobank software. Data are checked before final validation that displays the main information (Table 1).

Results

Patient consent and thyroid specimens

From September 1, 2004, to October 31, 2009, 1544 patients underwent surgery for a thyroid disease; 1454 of 1544 (94%) thyroid specimens and derived samples were included in the biobank; 6% (90/1544) of thyroids was not included because (1) the size of the tumor lesion was less than 0.3 cm, (2) the delay between surgical resection and tissue management (cold ischemia) was considered to be too long (longer than 30 min), (3) the patient did not give consent (n = 2); 93% (1440/1544) of patients signed consents for inclusion of tissue and blood samples in the biobank (Fig. 3). Consent was obtained before surgery in 1380 of 1440 (96%) and retrospectively in 60 of 1440 (4%) patients. If thyroid specimens were not obtained despite the patient's consent, a feedback note was given to the patient to explain why a sample was not stored frozen. The percentage of frozen thyroid tissues included with signed consent forms obtained per year is shown in Fig. 3A. Thyroid tissues associated with the following pathologies were included in the biobank: malignant tumors (253 cases), benign tumors, including the so-called tumors of uncertain malignant potential (388 cases),¹² nodular hyperplasia (673 cases), Graves' disease (111 cases), and thyroiditis (29 cases) (Fig. 3B).

FIG. 3.

(A) Number of thyroids included in the biobank and number of signed consent forms per year from 2004 to 2009. Percentages represent the ratio of thyroid inclusion and obtained consent forms signed by patients. (B) Pathologies of thyroids included in the biobank.

Details of the different thyroid pathologies

The details of the malignant and benign pathologies are shown in Fig. 4A–C. For 34 of 1454 (2%) of cases, the potential infectious status was prospectively (22 cases) or retrospectively (12 cases) known. These include HIV (16 cases), hepatitis B or C (17 cases) seropositivities, and tuberculosis (1 case) disease; 2 of 1454 thyroids, which corresponded to mesenchymal (1 case) and lymphoma (1 case) tumors, were considered for inclusion in a different database following the INCa recommendations (www.e-cancer.fr). The age and sex of patients together with the different diseases are summarized in Table 2.

FIG. 4.

(A) Number of included thyroids according to the different malignant tumor diseases. C-PTC, conventional papillary thyroid carcinoma; FV-PTC, follicular variant of papillary thyroid carcinoma; MI-FC, minimally invasive follicular carcinoma; LI-FC, largely invasive follicular carcinoma. (B) Number of included thyroids according to the different benign tumor diseases. FT-UMP, follicular tumor of uncertain malignant potential; WDT-UMP, well-differentiated tumor of uncertain malignant potential. (C) Number of included thyroids according to the different inflammatory diseases. A color version of this figure is available in the online article at www.liebertonline.com.

Table 2.

Main Epidemiological Data

Pathology	Age (years) Average (± SD)	Male (%)	Female (%)
Follicular adenoma	53.6 (±14.35)	68 (32)	145 (68)
Oncocytic adenoma	54.0 (±15.00)	19 (22)	67 (78)
FT-UMP	55.0 (±18.20)	5 (36)	9 (64)
WDT-UMP	53.9 (±19.70)	4 (31)	9 (69)
Mesenchymal tumor	58.0	1 (100)	0 (0)
Papillary microcarcinoma	55.0 (±16.50)	1 (4)	23 (96)
C-PTC	46.8 (±13.70)	12 (16)	65 (84)
FV-PTC	48.1 (±13.90)	11 (29)	27 (71)
Other histological subtypes of PTC	52.9 (±20.80)	4 (36)	7 (64)
Minimally invasive follicular carcinoma	47.0 (±21.40)	3 (27)	8 (73)
Largely invasive follicular carcinoma	48.8 (±12.00)	2 (33)	4 (67)
Medullary thyroid carcinoma	59.3 (±13.30)	8 (44)	10 (56)
Poorly differentiated and undifferentiated carcinoma	73.0 (±10.80)	0 (0)	3 (100)
Lymphoma	77.5 (±5.00)	0 (0)	2 (100)
Hashimoto thyroiditis	46.0 (±12.80)	1 (20)	4 (80)
De Quervain thyroiditis	56.0 (±10.10)	1 (33)	2 (67)
Nonspecific lymphocytic thyroiditis	52.1 (±15.70)	1 (7)	13 (93)
Graves' disease	39.9 (±13.00)	76 (86)	12 (14)
Nodular hyperplasia	57.8 (±13.40)	116 (21.6)	433 (78.4)

FT-UMP, follicular tumor of uncertain malignant potential; WDT-UMP, well-differentiated tumor of uncertain malignant potential; C-PTC, conventional papillary thyroid carcinoma; FV-PTC, follicular variant of papillary thyroid carcinoma.

Frozen tissue specimens

Frozen tissue specimens were obtained during preparation of a frozen section for diagnosis (350/1454; 24%) or were prepared systematically (1104/1454; 76%). For most cases, at least 1 tissue specimen with a lesion and 1 specimen presumably without a lesion (confirmed by histological evaluation) were frozen. The average number of prepared cryotubes per surgical specimen of thyroidectomy was 14 (7 tubes for tumor or other lesional tissue and 7 tubes for presumably adjacent healthy tissues). For 200 of 1454 (14%) thyroids, only affected tissue specimens were frozen because normal thyroid tissue was not available. For 60 of 1454 (4%) surgical specimens, only presumably normal thyroid tissue was frozen because the affected area was too small to be taken for freezing. Finally, a total of 9967 specimens were frozen and the distribution according to the diagnosis is summarized in Table 3. The mean weight of the frozen specimens per pathology and per cryotube is listed in Table 4.

Table 3.

Number of Frozen Specimens

Number/Pathology	No. of lesional specimens	No. of corresponding nonlesional specimens	Total
Follicular adenoma	1025	976	2001
Oncocytic adenoma	434	386	820
FT-UMP	62	56	118
WDT-UMP	61	46	107
Papillary microcarcinoma	35	128	163
C-PTC	275	385	660
FV-PTC	173	200	373
Other histological subtypes of PTC	52	44	96
Minimally invasive follicular carcinoma	60	20	80
Largely invasive follicular carcinoma	36	23	59
Medullary thyroid carcinoma	59	81	140
Poorly differentiated and undifferentiated carcinoma	24	12	36
Lymphoma	11	7	18
Hashimoto thyroiditis	22	8	30
De Quervain thyroiditis	14	0	14
Nonspecific lymphocytic thyroiditis	60	9	69
Graves' disease	384	13	397
Nodular hyperplasia	2611	1875	4486
Total	5398	4269	9667

Table 4.

Weight of Frozen Tissue Specimens

Pathology/Weight (mg)	Lesional tissue (mg)	Corresponding nonlesional tissue (mg)
Follicular adenoma	459 (±498)	372 (±269)
Oncocytic adenoma	546 (±544)	363 (±241)
FV-UMP	314 (±293)	224 (±172)
WDT-UMP	466 (±363)	360 (±228)
Papillary microcarcinoma	118 (±160)	460 (±301)
C-PTC	170 (±200)	361 (±264)
FV-PTC	319 (±328)	371 (±227)
Other histological subtypes of PTC	427 (±361)	354 (±250)
Minimally invasive FC	670 (±780)	407 (±234)
Largely invasive FC	601 (±413)	260 (±237)
Medullary thyroid carcinoma	148 (±178)	230 (±189)
Poorly differentiated and undifferentiated carcinoma	747 (±429)	480 (±469)
Lymphoma	319 (±30)	612 (±131)
Hashimoto thyroiditis	235 (±298)	145 (±162)
De Quervain thyroiditis	502 (±667)	0
Nonspecific lymphocytic thyroiditis	340 (±361)	472 (±554)
Graves' disease	657 (±564)	416 (±72)
Nodular hyperplasia	545 (±577)	364 (±301)
General mean	492 (±528)	369 (±275)

Blood samples

Since September 2006, prospective collection of plasma and sera was set up in parallel to tissue collection for all patients undergoing thyroid surgery. For 600 of 1454 (41%) patients, blood samples are available; 1250 plasma and 873 serum aliquots have been currently stored.

Nucleic acid samples extracted from tissue specimens

The time between surgical thyroidectomy and freezing varied from 15 to 30 min (mean time: 23 min). RNA and DNA were extracted from 510 specimens from 510 patients. The quality of the DNA and RNA was controlled as described in the Patients and Methods section and as previously described.¹³ The quantity and quality of the extracted nucleic acids from a quite similar weight of tissue varied depending on the pathology. Overall, the quantity of extracted nucleic acids per 50 mg of thyroid tissue was higher in tumor tissues than in nontumor or normal tissues. However, quality was variable depending on the underlying disease. Thus, for example, lesional tissue showing an intense inflammatory infiltrate (thyroiditis or severe lymphocytic reaction associated with papillary thyroid carcinoma) had a lower RIN, between 4.5 and 6.9 (mean = 6.3), than those of the tumors associated with a low level of inflammatory and necrotic areas, which had an RIN comprised between 6.8 and 9.1 (mean = 7.5).

After having collected RNA and DNA systematically from each thyroid for 2 years, the extracts were made available for specific research projects. To check the quality of the different procedures set up for the biobank, the quality of the RNA of 1 of each 20 thyroids is systematically controlled nowadays.

TMA paraffin blocks

The selection of the areas of interest for building TMAs was made by one of the surgical pathologists (S.L., M.I., I.L., C.B., or P.H.) on the different sections stained with H&E. These different slides were stored. As a result, TMAs are available for the following diseases: papillary carcinoma (28 paraffin blocks), follicular carcinoma (26 paraffin blocks), tumor of uncertain potential malignancy (18 paraffin blocks), nodular hyperplasia (14 paraffin blocks), adenoma (28 paraffin blocks), Graves' disease (20 paraffin blocks), and thyroiditis (8 paraffin blocks). Each block includes between 45 and 90 cases (with both selected lesional and perilesional areas in triplicate).

Primary cultures of thyrocytes

The success rate in establishing cell lines from the primary cultures was high, as epithelial cells from only 4 of 39 thyroid specimens were not obtained for unknown reasons. We checked that cultured cells obtained from the thyroid biopsy specimens shared biological features of thyroid tumors in situ. In particular, we looked at the phenotype of these cells obtained from primary cultures by immunocytochemistry using antibodies against antigens expressed by neoplastic and/or normal thyrocytes (thyroglobulin, thyroid transcription factor-1, cytokeratin 19, galectin 3, HBME1). We obtained a similar staining profile for epithelial cells from both primary cultures and biopsies. We concluded that the phenotype of the cells in primary culture tolerate cryopreservation well. Thus, the primary culture of thyroid biopsy material should not affect the ability to preserve biospecimens for other research purposes.

Discussion

The rules and conditions for biobanking have changed dramatically in France since the beginning of 2000. This occurred for 2 main reasons: (1) the continual increase in the use of human cells and tissues and other biological material for translational research, and (2) rapid progress in biotechnological approaches.¹⁴ In this regard, “modern” biobanks are now dependent on different biotechnological methodologies.^15,16 It is essential to develop national and international cooperation and to harmonize quality controls to satisfy future needs. The information given for “best practices” for biobanking by some institutions or organizations such as the International Society for Biological and Environmental Repositories (http://www.isber.org) is of great interest.¹⁷ Although information concerning the diagnosis of many disorders can now be obtained from formalin-fixed, paraffin-embedded tissues using, for instance, PCR-based testing rather than depending solely on fresh or frozen tissue, the need for preservation of unfixed tissue and other biological material remains an important requirement for the development and validation of new molecular tests, including those ultimately designed to use fixed archived tissue.⁵ Thus, frozen samples provide a crucial source of biochemically unaltered nucleic acids and proteins for translational research designed to investigate the mechanisms of diseases.

Frozen-tissue banks are an important resource for molecular pathology laboratories. However, the development of a frozen-tissue bank requires implementation of very efficient methods for collecting and freezing samples of fresh abnormal and normal tissues as a part of the routine activity of surgical pathology laboratories. It is also necessary to preserve additional types of frozen samples, such as cultured tumor or nontumor cells, serum, or plasma samples as well as aliquots of isolated DNA and RNA. Moreover, the growing clinical application of molecular diagnostic techniques requires standardization of the methods used to freeze, store, and inventory biological samples.⁵ However, it is difficult to systematically perform all these procedures for all specimens in the course of the routine activity of a surgical pathology laboratory. Consequently, we strongly believe that it is best to initially focus these procedures on only selected pathologies and/or organs. Thus, the present study describes the experience of an academic institution (Pasteur Hospital) in setting up a biobank targeted mainly to thyroid pathology.

We focused on thyroid pathology for different reasons. First, the incidence of thyroid lesions, particularly of thyroid cancer, is currently increasing in different parts of the world, particularly in Europe and the United States.^18–21 It is noteworthy that despite this increase, and in comparison to other types of carcinomas, there are currently only few available collections of such tumors, particularly in France.⁶ We also took the opportunity to collect frozen and paraffin-embedded tissue (TMAs) and different biological fluid samples from patients with thyroid diseases, as thyroid surgery is a major activity of our institution and because of the high number of patients with diseases of the thyroid in our geographical area. From the pathologists' viewpoint, controversial aspects of thyroid diseases and interindividual variations in the diagnosis of certain lesions such as follicular well-differentiated thyroid lesions justify the development of new tools designed to improve the accuracy of diagnosis.^12,22–24 In addition, recent studies that concern the molecular mechanisms of thyroid tumorigenesis and future studies that aim to develop new approaches leading to more accurate diagnoses will probably be done using frozen biological specimens.^25–27 Finally, some thyroid cancers such as carcinomas refractory to conventional treatment and some medullary carcinomas are particularly aggressive. As they do not benefit from satisfactory therapeutic approaches, new targeted therapies are currently being developed.^28,29 In this regard, targeted therapies will put into practice the discoveries of molecular mechanisms of carcinogenesis.^30,31

The thyroid biobank of the Pasteur Hospital is made up of different biological collections obtained from the same cohort of patients. These collections were established in compliance with criteria defined by INCa in 2006.⁷ Although most grants are targeted to projects into cancer research, particularly in France, it is also important to collect nontumor tissue from the same organ, because many interesting translational research projects can be developed for benign, dystrophic, autoimmune, and inflammatory processes that arise in the thyroid gland. Thus, we have set up a biobank that includes tumor and nontumor diseases.⁹

In the last few decades, an increasing number of human biobanks have been set up around the world, which underlies the usefulness of obtaining different biological products for human clinical and translational research.^32–34 Before beginning the current project, we questioned whether it was still worth creating yet another biobank. Finally, we concluded that it was necessary to go in this direction for different reasons. Some previously established collections may not remain available for research, as they cannot fulfill some of the criteria and standards for biobanking that were established later.^35–39 Although there are many scientific justifications for the creation of tissue and DNA databanks, the storage and use of human biological materials continue to face legal dilemmas and stimulate discussion.^40–49 It is necessary to obtain for clinical research a consent form signed by the patient. Moreover, complete epidemiological and clinical data, follow-up of the patient, pathological data, digitalized images corresponding to frozen tissue specimens, and data concerning the quantity and the integrity of nucleic acids are also necessary.

With the widespread use of nucleic acids and proteins, hundreds of biomarkers are in need of validation in cohorts of well-annotated clinical samples. TMAs are emerging as the tool for performing rapid protein expression analyses on a large number of clinical samples.⁵⁰ As TMAs represent an increasingly validated means of understanding the clinical impact of diagnosis-, prognosis-, and therapy-related markers, we decided to prospectively set up TMAs that are representative of different thyroid pathologies.

Attention must be paid to different safety procedures. The strict confidentiality of all clinical and pathological data must be respected. This should allow the use of human biological products for targeted projects and help the transfer of collected tissue to other countries. Except for some rare lymphomas or sarcomas arising in the thyroid, all collected tissues in our thyroid biobank qualify for use in clinical and translational research.⁵¹ Before giving access to human thyroid samples, each project is evaluated by a scientific committee. All projects are submitted to the ethics committee at our institution. The different procedures used before transfer of material for translational research are listed in Fig. 5.

FIG. 5.

Different steps from the submission of a project for translational research to the discharge of samples for performance of the project. *See criteria for inclusion of samples in Table 1. MTA, material transfer agreement.

All procedures for setting up the thyroid biobank have been made by a permanent and professional staff that is specifically dedicated to this activity: technicians working with the pathologists in freezing and storing samples; technicians for nucleic acid extraction and control of integrity; technicians for the plasma, sera, and primary culture bank; technicians for data and quality control management; a secretary for registration and procedures; and a pathologist to control the different histological images (www.biobank06.com/annual). The biobank is physically located in the Department of Surgical Pathology (www.biobank06.com; center of biological resources/partners) and its operation is supervised by a surgical pathologist who oversees the scientific and administrative aspects.

The usefulness of establishing such a biobank targeted to a single pathology needs to be evaluated in the future. Different indicators will rapidly help check if setting up this large biobank is of interest. In particular, these indicators will include the evaluation of the ratio between the number of stored and used samples, the number and impact factor of publications where the name of the biobank is mentioned, the number and quality of scientific collaborations, and the number of samples sent and sold to academic and private partners. Estimation of the cost of human biological resources is difficult and depends in particular on the different facilities set up in the biobank and on the expertise area of the biobank. As an example, for academic partners, we used the costs recommended by the INCa in France (www.e-cancer/recherche/les-ressourcesbiologiques). Finally, a biobank of high standing could become a cornerstone in large translational consortia. In this regard, the thyroid collection of the biobank will be included in different programs of translational research currently sponsored by the INCa. Moreover, it is probable that such a thyroid biobank will benefit patients, by exploring different new molecular events.

There are many reports of the modus operandi of population-based and disease-specific research biobanks. These publications reveal both the diversity of the approaches and some common difficulties. The cost of establishing a human biological collection is high. It is thus necessary to establish collaboration with academia and industry and network with different biobanks working on similar targeted pathologies. For this reason, this biobank participates in the local canceropole Provence Alpes Côte d'Azur (PACA) consortium (http://www.canceropole-paca.com) as well as in the current European biobanking project.⁸ We strongly believe that the development of a human biobank requires implementation of efficient methods for collecting and freezing samples of fresh tumor and normal tissues and biological fluids as a part of the routine activity of surgical pathology laboratories. Moreover, the effort made will only be productive if biobanking and its associated computing systems encourage the translational bridge linking new molecular information to its clinical significance.^52,53 Finally, it should be kept in mind that the setup of a biobank of high quality for different types of human biological materials ultimately aims to satisfy the interests of the patient community.⁵⁴ In this regard, the integration of biological resources from well-organized biobanks with powerful molecular and “omics” approaches and bioinformatic tools hold promise in advancing further our knowledge of disease development, thus leading to better prevention and treatment strategies.

Footnotes

Acknowledgments

The authors thank the Canceropole PACA (Clara Ducord) and INCa (Véronique Atger and Pascal Boucher) as well as Laurent Jacotot and Philippe Vaglio for helping to integrate the thyroid collection into the Biobank-PACA ().

Author Disclosure Statement

No competing financial interests exist.

References

Bova

, Eltoum

, Kiernan

et al. Optimal molecular profiling of tissue and tissue components: defining the best processing and microdissection methods for biomedical applications. Mol Biotechnol, 2005; 29:119–152.

Hofman

. DNA microarrays. Nephron Physiol, 2005; 99:85–89.

Medeiros

, Riql

, Anderson

et al. Tissue handling for genome-wide expression analysis: a review of the issues, evidence, and opportunities. Arch Pathol Lab Med, 2007; 131:1805–1816.

Florell

, Coffin

, Holden

et al. Preservation of RNA for functional genomic studies: a multidisciplinary tumor bank protocol. Mod Pathol, 2001; 14:116–128.

Hofman

, Ilie

, Gavric-Tanga

et al. Role of the surgical pathology laboratory in the pre-analytical approach of molecular biology techniques. Ann Pathol, 2010; 30:85–93.

Chabannon

, Penault-Llorca

, Cambon-Thomsen

et al. Etat des lieux des tumorothèques françaises au premier trimestre de l'année 2005. Bull Cancer, 2006; 93:S221–S228.

Copin

, Bibeau

, Durand

et al. Visibilité nationale des tumorothèques et cadre d'interopérabilité. Principes généraux et critères standarts de description des échantillons cryopréservés. Bull Cancer, 2006; 93:S241–S245.

Yuille

, Van Ommen

, Bréchot

et al. Biobanking for Europe. Brief Bioinform, 2008; 9:14–24.

Hofman

. La tumorothèque/tissuthèque CHU-CRLCC-UNSA de la région niçoise. Med Sci, 2006; 22:21–25.

10.

Hofman

, Bonnetaud

, Gaziello

et al.

The Nice CHU biobank experience to collect patients' informed consent for research context (2004–2009)

Ann Pathol, 2010; 30:337–343.

11.

Van Staveren

, Weiss Solis

, Delys

et al. Gene expression in human thyrocytes and autonomous adenomas reveals suppression of negative feedbacks in tumorigenesis. Proc Natl Acad Sci U S A, 2006; 103:413–418.

12.

Hofman

, Lassalle

, Bonnetaud

et al. Thyroid tumours of uncertain malignant potential: frequency and diagnostic reproducibility. Virchows Arch, 2009; 455:21–33.

13.

Lassalle

, Hofman

, Marius

et al. Assessment of morphology, antigenicity, and nucleic acid integrity for diagnostic thyroid pathology using formalin substitute fixatives. Thyroid, 2009; 19:1239–1248.

14.

Becich

. The role of the pathologist as tissue refiner and data miner: the impact of functional genomics on the modern pathology laboratory and the critical roles of pathology informatics and bioinformatics. Mol Diagn, 2000; 5:287–299.

15.

Qualman

, Bowen

, Brewer-Swartz

et al. The role of tumor banking and related informatics in molecular research. Ladanyi

, Gerald

. Expression Profiling of Human Tumors: Diagnostic and Research Applications. Totowa, NJ: Humana Press, Inc., 2003; 103–117.

16.

Naber

. Continuing role of a frozen-tissue bank in molecular pathology. Diagn Mol Pathol, 1996; 5:253–259.

17.

Pitt

. 2008 Best practices for repositories. Collection, storage, retrieval and distribution of biological materials for research. International Society for Biological and Environmental Repositories. Cell Preservation Technol, 2008; 6:1–58.

18.

Alston

, Geraci

, Eden

et al. Changes in cancer incidence in teenagers and young adults (ages 13 to 24 years) in England 1979–2003. Cancer, 2008; 113:2807–2815

19.

Davies

, Welch

. Increasing incidence of thyroid cancer in the United States, 1973–2002. JAMA, 2006; 295:2164–2167.

20.

Colonna

, Grosclaude

, Remontet

et al.

Incidence of thyroid cancer in adults recorded by French cancer registries (1978–1997)

Eur J Cancer, 2002; 38:1762–1768.

21.

Pacini

, Castagna

, Brilli

et al. Differentiated thyroid cancer: ESMO clinical recommendations for diagnosis, treatment and follow-up. ESMO Guidelines Working Group. Ann Oncol, 2008; 19,Suppl 2:ii99–101.

22.

Rosai

. Handling of thyroid follicular patterned lesions. Endocr Pathol, 2005; 16:279–283.

23.

Serra

, Asa

. Controversies in thyroid pathology: the diagnosis of follicular neoplasms. Endocr Pathol, 2008; 19:156–165.

24.

Suster

. Thyroid tumors with a follicular growth pattern: problems in differential diagnosis. Arch. Pathol. Lab Med, 2006; 130:984–988.

25.

DeLellis

, Lloyd

, Heitz

, Eng

. Pathology and Genetics. Tumours of Endocrine Organs. World Health Organisation Classification of Tumors. Lyon, France: IARC Press, 2004.

26.

Nikiforov

. Thyroid carcinoma: molecular pathways and therapeutic targets. Mod Pathol, 2008; 21,Suppl 2:S37–S43.

27.

Sobrinho-Simoes

, Preto

, Rocha

et al. Molecular pathology of well-differenciated thyroid carcinoma. Virchows Arch, 2005; 447:787–793.

28.

Etit

, Faquin

, Gaz

et al. Histopathologic and clinical features of medullary microcarcinoma and C-cell hyperplasia in prophylactic thyroidectomies for medullary carcinoma: a study of 42 cases. Arch Pathol Lab Med, 2008; 132:1767–1773.

29.

Neff

, Farrar

, Kloos

et al. Anaplastic thyroid cancer Endocr. Metab Clin North Am, 2008; 37:525–538.

30.

Puxeddu

, Durante

, Avenia

et al. Clinical implications of BRAF mutation in thyroid carcinoma. Trends Endocrinol Metab, 2008; 19:138–145.

31.

Smallridge

, Marlow

, Copland

. Anaplastic thyroid cancer: molecular pathogenesis and emerging therapies. Endocr Relat Cancer, 2008; 16:17–44.

32.

Auray-Blais

, Patenaude

. A biobank management model applicable to biomedical research. BMC Med Ethics, 2006; 7:1–9.

33.

Goebell

, Groshen

, Schmitz-Dräger

et al. Concepts for banking tissue in urologic oncology. Int Bladder Cancer Bank, 2005; 11:413–415.

34.

Thomas

, Williams

, Becker

et al. Creation of a tumour bank for post Chernobyl thyroid cancer. Clin Endocrinol (Oxf), 2001; 55:423.

35.

Grizzle

, Grody

, Noll

et al. Recommended policies for uses of human tissue in research, education, and quality control. Ad Hoc Committee on Stored Tissue, College of American Pathologists. Arch Pathol Lab Med, 1999; 123:296–300.

36.

Jewell

, Srinivasan

, McCart

et al. Analysis of the molecular quality of human tissues: an experience from the cooperative human tissue network. Am J Clin Pathol, 2002; 118:733–741.

37.

Teodorovic

, Therasse

, Spatz

et al. Human tissue research: EORTC recommendations on its practical consequences. Eur J Cancer, 2003; 39:2256–2263.

38.

Trouet

. New European guidelines for the use of stored human biological materials in biomedical research. J Med Ethics, 2004; 30:99–103.

39.

Von Versen

, Mönig

, Salai

et al. Quality issues in tissue banking: quality management systems- a review. Cell Tissue Bank, 2000; 1:181–192.

40.

Anderson

, Schonfeld

. Patient consent in the era of de-identified research databases. J Clin Oncol, 2006; 24:720–721.

41.

Elger

, Caplan

. Consent and anonymization in research involving biobanks: differing terms and norms present serious barriers to an international framework. EMBO Rep, 2006; 7:661–666.

42.

Furness

, Nicholson

. Obtaining explicit consent for the use of archival tissue samples: practical issues. J Med Ethics, 2004; 30:561–564.

43.

Godard

, Schmidtke

, Cassiman

et al. Data storage and DNA banking for biomedical research: informed consent, confidentiality, quality issues, ownership, return of benefit. A professional perspective. Eur J Hum Genet, 2003; 11:S88–S122.

44.

Hansson

, Dillner

, Bartram

et al.

Should donors be allowed to give broad consent to future biobank research?

Lancet Oncol, 2006; 7:266–269.

45.

Malone

, Catalano

, O'Dwyer

et al. High rate of consent to bank biologic samples for future research: the Eastern Cooperative Oncology Group experience. J Natl Cancer Inst, 2002; 94:769–771.

46.

Maschke

, Murray

. Ethical issues in tissue banking for research: the prospects and pitfalls of setting international standards. Theor Med Bioeth, 2004; 25:143–155.

47.

Maschke

. Alternative consent approaches for biobank research. Lancet Oncol, 2006; 7:193–194.

48.

Nilstun

, Hermeren

. Human tissue samples and ethics-attitudes of the general public in Sweden to biobank research. Med Health Care Philos, 2006; 9:81–86.

49.

Sebire

, Dixon-Woods

. Towards a new era of tissue-based diagnosis and research. Chronic Illness, 2007; 3:301–309.

50.

Camp

, Neumeister

, Rimm

. A decade of tissue microarrays: progress in the discovery and validation of cancer biomarkers. J Clin Oncol, 2008; 26:5630–5637

51.

Coindre

, Sigaux

, Delsol

. Tumorothèques à visée sanitaire: définition, intérêts et recommendations. Bull Cancer, 2006; 93:S213–S220.

52.

Meade

. The future of biobank. Lancet, 2003; 362:492.

53.

Morente

, Alonso

. Current challenges of human tumour banking. Hematol Oncol, 2005; 23:54–56.

54.

Oosterhuis

, Coebergh

, Van Veen

. Tumour banks: well-garded treasures in the interest of patients. Nat Rev Cancer, 2003; 3:73–77.

Setting up a Prospective Thyroid Biobank for Translational Research: Practical Approach of a Single Institution (2004–2009,Pasteur Hospital,Nice,France)

Abstract

Introduction

Patients and Methods

Ethics, consent, informative procedures, and confidentiality

Clinical and epidemiological data and patient follow-up

Tissue specimen management

Pathological data

Tissue microarrays

Blood sample management

Primary culture of thyrocytes

Quality assessment and nucleic acid collection

Storage facilities and safety procedures

Information technology

Results

Patient consent and thyroid specimens

Details of the different thyroid pathologies

Frozen tissue specimens

Blood samples

Nucleic acid samples extracted from tissue specimens

TMA paraffin blocks

Primary cultures of thyrocytes

Discussion

Footnotes

Acknowledgments

Author Disclosure Statement

References