Rapid advances are being made in cancer drug therapy. Since molecularly targeted therapy has been introduced, personalized medicine is being practiced, pathological tissue from malignant tumors obtained during routine practice is frequently used for genomic testing. Whereas cytological specimens fixed mainly in alcohol are considered to be more advantageous in terms of preservation of the nucleic acid quality and quantity. This article is aimed to share the information for the proper handling of cytological specimens in practice for genomic medicine based on the findings established in “Guidelines for Handling of Cytological Specimens in Cancer Genomic Medicine (in Japanese)” published by the Japanese Society of Clinical Cytology in 2021. The three-part practical guidelines are based on empirical data analyses; Part 1 describes general remarks on the use of cytological specimens in cancer genomic medicine, then Part 2 describes proper handling of cytological specimens, and Part 3 describes the empirical data related to handling of cytological specimens. The guidelines indicated proper handling of specimens in each fixation, preparation, and evaluation.

© 2023 The Author(s). Published by S. Karger AG, Basel

Introduction

Rapid advances are being made in cancer drug therapy. Since molecularly targeted therapy has been introduced, personalized medicine is being practiced [1, 2]. In personalized medicine, gene mutations present in cancer cells are examined on a case-by-case basis to discover drugs effective against the mutations [3]. In Japan, gene panel tests aimed at examining multiple gene mutations simultaneously are covered under national health insurance since June 2019, and expert panels to discuss test results are held on a daily basis. Cancer genomic medicine now appears to be fully incorporated into routine clinical practice. Under such circumstances, the Japanese Society of Clinical Cytology (JSCC) has decided to publish the “Guidelines for Handling of Cytological Specimens in Cancer Genomic Medicine” regarding cytological specimens suitable for use in cancer genome diagnostics. The aims of this project are to improve and standardize the reporting of cytopathology, improve communication between cytopathologists and clinicians, improve patient care, focus on the current and potential role of cytopathology at these sites.

Currently, in cancer genomic medicine, nucleic acid sources are tissues that have been fixed in formalin and embedded in paraffin [4]. However, while experience in performing gene panel testing is increasing, good results cannot be obtained in some cases wherein formalin-fixed, paraffin-embedded (FFPE) specimens are stored for too long or not fixed properly. As cancer genomic medicine is already a part of routine clinical practice, pathological specimens nowadays are required to be suitable not only for morphological characterization but also for molecular-level characterization, i.e., gene mutation analysis. This also applies to cytological specimens. Compared with paraffin-embedded tissue specimens, cytological specimens fixed mainly in alcohol are considered to be more advantageous in terms of preservation of the nucleic acid quality [5]. Therefore, cytological specimens represent a promising class of nucleic acid sources in gene panel testing. Nevertheless, taking precautions at the pre-analytical stage is definitely required, including the use of an appropriate fixative and proper fixing time.

In order to make the guidelines applicable to genomic and cytologic studies, JSCC created the Working Group on cytology in the era of cancer genomic medicine and a subcommittee thereof, the Working Group on quality control of specimen processing for respiratory cytology in the genome era. These working groups first conducted a questionnaire survey on the usage of cytological specimens keeping the era of cancer genomic medicine in mind and are now conducting empirical experiments involving several laboratories for identifying all matters related to the use of cytological specimens in cancer genomic medicine that require attention. Here, we provided a lot of important matters, such as preservation of samples and extraction methods of nucleic acid. Now we are still investigating several matters, such as the effect of various methods of cell block preparation on the quality of nucleic acid, with empirical experiments, and we plan to publish the results on the website of JSCC as reportable data are obtained in order to share the information with members [6]. Depending on the methods used, the purpose of cancer genomic medicine may not be fully met; however, we hope that the problems identified and solutions explored in these empirical experiments will contribute to the development of new methods and solutions to problems. Therefore, we translated the guideline to English version for international usage.

Part 1: Use of Cytological Specimens in Cancer Genomic Medicine

In Japan, cancer gene panel testing with the use of next-generation sequencing (NGS) has been approved for clinical use in solid cancer cases and is covered under insurance since June 2019. This multiplex genetic testing is divided into companion diagnostic tests and genomic profiling tests in clinical practice. The former aims to access standard treatments for which established evidence is available, and NGS-based tests are currently used primarily in the field of lung cancer. On the other hand, the latter has the primary aim of access to investigational treatments for which evidence is yet to be established through research and primarily development and covers all solid cancers. Since the clinical use of NGS-based tests has started as insurance-covered medical services, many issues on operation with a clinical system distinct from the conventional one have emerged. In particular, FFPE tissue samples, which are mainly used for oncogene panel testing, are unusable or sometimes provide unanalyzable results depending on the specimen processing methods and FFPE block storage conditions employed [7]. As a result, expectations related to the use of cytological specimens are increasing.

Regarding the usage of cytological specimens in conventional gene-based testing, in pathogen testing, human papillomavirus testing using gynecological samples is the most advanced, and many human papillomavirus testing systems have been approved as in vitro diagnostics and covered by health insurance. Meanwhile, cancer gene testing is used for limited purposes such as lung cancer companion diagnostic tests using respiratory specimens. According to the “Survey on the Usage of Cytological Specimens” conducted by JSCC’s Working Group on Cytology in the Era of Cancer Genomic Medicine in April 2019 (see online suppl. Text, Survey Information; for all online suppl. material, see www.karger.com/doi/10.1159/000528346), FFPE tissue samples were mainly used for somatic gene tests conducted as a companion diagnostic method as per all survey items, and virtually no cytological specimens were used in Human Epidermal Growth Factor Receptor 2 (HER2)/Erb-B2 Receptor Tyrosine Kinase 2 (ERBB2) fluorescent in situ hybridization (FISH) testing for breast and gastric cancers, Rat sarcoma viral oncogene (RAS) mutation testing for colorectal cancers, and microsatellite instability testing for solid cancers (see online suppl. Text. Survey Data). In lung cancer diagnostics, cytological specimens were used for Epidermal growth factor receptor (EGFR) mutation, anaplastic lymphoma kinase (ALK) FISH, ROS proto-oncogene 1 (ROS1) fusion gene, and BRAF proto-oncogene (BRAF) mutation tests; however, the majority of them used FFPE cell block samples. Except for FFPE cell block samples, EGFR mutation tests, such as the peptide nucleic acid-locked nucleic acid polymerase chain reaction (PCR) Clamp method, have been commonly applied for cytological specimens for many years as laboratory-developed test (LDT) methods that can be included in insurance claims. ROS1 fusion gene testing, which is the first ribonucleic acid (RNA)-based companion diagnostic test, approved for the use of cytological specimens in Japan. In addition, the FISH test using urinary specimens (cells in urine) and performed as a diagnostic aid for the recurrence of bladder cancer is approved as an in vitro diagnostics and covered by insurance; however, its use is limited.

While the use of single-gene testing has been established, NGS-based multiplex companion diagnostic and genomic profiling tests were included in the insurance coverage in 2019, and cancer genomic medicine has been fully implemented in clinical practice since then. FFPE tissue samples are mainly used in these tests; however, cytological specimens such as FFPE cell blocks can also be used. The use of cytological specimens and research for their use are advanced overseas [8-10]; various forms of cytological specimens, such as FFPE cell block, conventional smear, and liquid-based cytology (LBC) specimens, are used [11, 12]. In cancer genomic medicine, histological and cytological specimens are complementary to each other, and it is important to understand their characteristics in an orderly manner (Table 1).

Table 1.

Comparison of the characteristics of histological and cytological specimens

As a result of various improvements in cytological specimen processing at each institution, multiple procedures are currently being used in Japan. For cancer gene testing, cytological specimens are relatively frequently used in lung cancer companion diagnostic tests such as the EGFR mutation testing described above; however, conventional single-gene tests are conducted without making careful considerations regarding the tumor cell (TC) content of cytological specimens. A TC content of ≥20% is usually required for conducting NGS-based gene panel tests to achieve adequate test performance, and estimation of this content is an important process. Current cancer gene panel tests, both companion diagnostic and genomic profiling tests, are primarily used for selecting molecularly targeted therapies, and tests conducted with samples having variable nucleic acid quality due to interinstitutional differences in specimen processing or with inappropriate specimens can provide incorrect test results, causing patients to lose treatment opportunities, which is a great disadvantage. In cancer genomic medicine wherein multigene test information is required, gene panel tests that use “gene panels” covering multiple genes require more rigorous sample handling than single-gene tests to obtain the required quality of results. While mainly FFPE tissue samples are used in Japan, advanced medical services that employ gene panel testing systems using cytological specimens are being provided in 2020. Moreover, the use of gene panel testing systems, which have currently gained approval for the use of FFPE tissue samples, is going to increase in the near future. Therefore, there is an urgent need to standardize the pre-analytical and analytical phases of their tests carried out by medical institutions.

Part 2: Proper Handling of Cytological Specimens

In this guideline, appropriate and standard handling methods are described that may be required by general medical institutions for the widespread use of cytological specimens in the future and to promptly respond to the introduction of cancer genomic medicine in routine clinical practice in the field of cytodiagnosis. A wide variety of cytological specimen processing methods are used currently. This guideline focuses on (i) the processing and preparation methods of cytological specimens expected to be used in cancer genomic medicine, such as oncogene panel testing in particular [12], and (ii) methods for determining test suitability according to TC content evaluation (items in bold in Fig. 1).

Fig. 1.

Major cytological specimens/specimen types used in cancer genomic medicine. Items in bold showed this guideline focuses; the processing and preparation methods of cytological specimens expected to be used in cancer genomic medicine, and the methods for determining test suitability according to TC content evaluation.

Grades of recommendations based on evidence-based medicine as used in clinical practice guidelines are not listed in this guideline. Instead, guideline’s recommendations that are based on the empirical data and literature information described later are described in the following two categories: clinical recommendations (C) and research recommendations (R). The former is items recommended as best practices in routine clinical practice. The latter is items recommended considering the usage for genomic screening-based intervention studies (highly comprehensive genomic analysis) or preliminary clinical practice not covered by health insurance. Note that these recommendations need to be revised and updated in a timely manner based on future developments in genomic diagnosis.

Handling of Specimens before ProcessingHandling of Specimens after Specimen Collection to Processing Initiation

Processing of cytological specimens collected by preparing smears, scraping, fine-needle aspiration, etc., should be initiated as soon as possible (C) (Fig. 2).

Fig. 2.

Effects of storage time and temperature on cell suspension specimens prepared in saline. DIN value of the specimens stored in a refrigerator for 3 days decreased mildly and did not differ significantly from the DIN value of the specimens stored at room temperature for 24 h. The DIN value of the specimens stored for 3 days at room temperature decreased compared with that of the specimens stored for 24 h at room temperature.

It is desirable to store cytological specimens suspended in physiological saline in a refrigerator (4°C), and they must be subjected to nucleic acid extraction as soon as possible (C) (Fig. 2).

LBC specimens should be treated with an LBC preservative solution as soon as possible after collection (C) (Fig. 2).

It is desirable to store LBC samples at room temperature or in a refrigerator (4°C), and they must be subjected to nucleic acid extraction promptly (C) (Fig. 2).

It is desirable to use liquid nitrogen or the dry ice acetone method for snap freezing of fresh samples. Frozen specimens should be stored at −180°C to −80°C and promptly subjected to nucleic acid extraction (R) [13].

Avoid leaving unprocessed cytological specimens at room temperature (approximately 25°C) whenever possible (N) (Fig. 2).

Avoid storing LBC samples in places with direct sunlight, high temperature, and high humidity (N).

For frozen specimens, avoid repeating thaw-freeze cycle (N) [13].

Other Points to Note

Prepare an organized working environment to prevent the cross-contamination (C) [13].

When specimens are collected from multiple tumors or different sites within tumors, identification codes should be provided for each sample (C) [13].

For liquid specimens such as body cavity fluids (e.g., pleural effusion, ascites) and urine, it is desirable to select an appropriate specimen processing method depending on specimen volume and cell content (C).

It is desirable to confirm the presence of target cells on microscope; however, when the specimen quantity is insufficient, it is desirable to set an in-house consensus whether to prioritize morphological or genomic diagnosis (C).

When specimens are available in sufficient tissue/TC quantities, concomitant use of multiple storage methods may be considered (R).

Supplementary Explanations

1S It is recommended to perform nucleic acid extraction as soon as possible because the deoxyribonucleic acid (DNA) quality tends to decrease as the storage duration increases in empirical experiments using cytological specimens suspended in physiological saline. If the specimen processing procedure cannot be initiated immediately, quality deterioration can be decelerated by storing the specimens in a refrigerator (4°C) instead of storing them at room temperature (approximately 25°C) [14].

1S Results of validation experiments on effects of the duration and temperature of cytological specimen storage tended to be similar to those of empirical experiments using histopathological specimens, and cytological specimens are recommended to be processed as quickly as histological specimens [13].

4S Refrigeration or frozen storage has been long known to decelerate degradation and putrefaction caused by enzymes and microorganisms. If additional processing of residual specimens using formalin-containing LBC fixatives cannot be done promptly, temporary storage in a refrigerator is recommended to reduce DNA fragmentation risk (C) [14]. It is also possible to carry out genetic testing after frozen storage [15]. Regarding the nucleic acid quality of unprocessed cytological specimens stored for a long time at an ultralow temperature and cytological specimens treated with an LBC preservative solution, it has been shown that certain yield/quality of DNA and RNA can be obtained after storage for at least about 6 months (Fig. 4, 5).

13S While there are multiple LBC preservative solutions with different compositions, specimen processing methods for cancer genomic testing have not been established currently. Therefore, when sufficient quantities of specimens can be obtained, it may be considered to use multiple storage methods concomitantly (R).

12S Because different specimen quality criteria are required depending on whether the specimen is intended for single-gene or gene panel testing, a method appropriate for the intended purpose should be chosen. For panel testing, for example, a method that enables the evaluation of TC content should be chosen because it is one of the criteria, whereas for single-gene testing for known mutations such as EGFR mutations, genetic test results may be useful for planning treatment strategies even with cerebrospinal fluid and pleural effusion specimens in which TCs cannot be confirmed. Therefore, it is desirable that the testing method is selected through the collaboration of clinicians and pathology/genetic testing laboratories.

13S Types of preservatives and nucleic acid extraction methods are still being standardized, and the optimum method has not been established yet. Thus, multiple methods may be used in parallel; however, a limited number of institutions can implement such an approach because it is laborious and time-consuming and requires a place for storage.

LBC SpecimensFor Each LBC Specimen

1. Collect a necessary and sufficient number of nucleated cells to conduct genomic analysis (C) [13].

2. The collected cells are immediately suspended and fixed in an LBC preservative solution (C).

3. For fixation duration, follow the LBC preservative solution’s respective manufacturer’s instruction manual (C).

4. Preferred sample preparation methods enable easy morphological evaluation of the distribution and percentage of TCs (C).

5. Confirm TC content using methods such as Papanicolaou staining and immunocytochemical staining (C) (online suppl. Fig. 1).

Nucleic Acid Extraction from Specimens Preserved Using LBC Preservative Solution Containing Formalin

6. When using a commercially available DNA extraction kit to extract DNA from FFPE tissue specimens preserved using an LBC preservative solution, proteolysis with a proteinase K solution should be performed in an appropriate condition because it affects nucleic acid yield and quality (C) [16] (Fig. 3).

Fig. 3.

Optimal nucleic acid extraction methods for specimens preserved using different LBC preservative solutions. a Cells of a lung adenocarcinoma cell line were fixed in an LBC preservative solution (CytoRichTM Red), and DNA was extracted with a commercial DNA extraction kit for FFPE tissues QIAamp® DNA FFPE Tissue Kit (Qiagen). b Cells were fixed in an LBC preservative solution CytoRichTM Blue (for nongynecological use [BD]), and DNA was extracted with QIAamp® DNA FFPE Tissue Kit. c Cells were fixed in CytoRichTM Red, and DNA was extracted with Maxwell® FFPE Plus DNA Kit (Promega). d Cells were fixed in CytoRichTM Blue, and DNA was extracted with Maxwell® FFPE Plus DNA Kit.

7. When using a commercially available DNA extraction kit to extract DNA from FFPE tissue specimens preserved using an LBC preservative solution, cross-links between nucleic acids and formalin should be removed by prolonged heating in an appropriate condition because it results in the deterioration of the nucleic acid quality (R) (Fig. 3).

Nucleic Acid Extraction from Specimens Preserved Using LBC Preservative Solution Not Containing Formalin

8. When using a commercially available DNA extraction kit to extract DNA from FFPE tissue specimens preserved using an LBC preservative solution, proteolysis with a proteinase K solution can be performed in a shorter time than that stated in the kit protocol (R) (Fig. 3).

9. When using a commercially available DNA extraction kit to extract DNA from FFPE tissue specimens preserved using an LBC preservative solution, removal of cross-links between nucleic acids and formalin by heating may not be performed (R) (Fig. 3).

10. Avoid using LBC specimens stored for more than about 6 months for genomic testing whenever possible (N).

Supplementary Explanations

1S It is recommended to collect a necessary and sufficient number of nucleated cells for genomic analysis and combine the conventional smear preparation with an evaluation of the distribution and percentage of TCs (C).

2−3S It is necessary to familiarize oneself with the cell morphologies of LBC specimens since they differ depending on which company’s LBC is used (R) (online suppl. Fig. 2).

3S LBC preservative solutions containing methanol or ethanol as the base component are referred to as methanol- or ethanol-based preservative solutions, whereas LBC preservative solutions containing a minute amount of formalin are referred to as formalin-based preservative solutions. Different types of LBC preservative solutions have different effects on nucleic acid quality (online suppl. Fig. 4–7). Cell blocks can be prepared from LBC preservative solutions, but it should be noted that LBC preservative solutions from different manufacturers differ in their composition and hemolytic effect [17, 18] (online suppl. Fig. 2 and 3).

2S Avoid the paucity of cell mass, cell collection from inappropriate sites, and cell degeneration caused by methodological differences in hemolysis and fixation as much as possible (N).

4−5S Papanicolaou staining is the standard method used for evaluating the distribution of TCs; however, additional evaluation using immunocytochemical or special staining is recommended (C).

6−7S Specific examples of formalin-containing LBC preservative solutions include CytoRichTM Red preservative solution (BD).

6S The protocol for QIAamp® DNA FFPE Tissue Kit (Qiagen) states “at 56°C for 1 hour or until the sample is completely dissolved.” The yield and quality can be ensured at 56°C for 30 minutes as long as the sample is completely dissolved (R) [16, 19] (Fig. 3). The yield and quality will be substantially reduced if this procedure is omitted (N) (Fig. 3).

6S The protocol for Maxwell® FFPE Plus DNA Kit (Promega) states, “incubate at 70°C for 1 hour or overnight.” When LBC-preserved specimens are used, incubation for 1–4 hours is sufficient (R) (Fig. 3). Effects of overnight incubation on the yield and quality have not been tested. However, they will be substantially reduced if this procedure is omitted (N) (Fig. 3).

7S The protocol for QIAamp® DNA FFPE Tissue Kit (Qiagen) states, “incubate at 90°C for 1 hour.” When specimens preserved using LBC preservative solutions are used, the yield and quality can be expected to be improved by shortening the incubation duration to 10 minutes (R) (Fig. 3). The yield and quality may be reduced if this procedure is omitted (N) [20, 21] (Fig. 3).

8−9S Specific examples of LBC preservative solutions not containing formalin include CytoRichTM Blue preservative solution (BD), ThinPrep® PreservCyt solution (Hologic), and Cellprep® vials (Roche).

8S The protocol for QIAamp® DNA FFPE Tissue Kit (Qiagen) states, “at 56°C for 1 hour or until the sample is completely dissolved.” The yield and quality can be ensured at 56°C for 30 minutes as long as the sample is completely dissolved (R). This procedure can be omitted in some cases; however, it is better not to, given the original role of proteolytic enzymes (Fig. 3).

8S The protocol for Maxwell® FFPE Plus DNA Kit (Promega) states “incubate at 70°C for 1 hour to overnight.” When specimens preserved using LBC preservative solutions are used, incubation for 30 min is sufficient (R). Incubation up to 4 h is considered to have no effects on the yield and quality (R) (Fig. 3). Effects of overnight incubation on the yield and quality have not been tested. This procedure can be omitted in some cases; however, given the original role of proteolytic enzymes, it is better not to.

8S Specimens preserved in LBC preservative solutions not containing formalin (CytoRichTM Blue preservative solution) may be incompatible with the magnetic particle method (Fig. 3). Currently, conducting sufficient tests before use is advised.

9S While the protocol for QIAamp® DNA FFPE Tissue Kit (Qiagen) states “incubate at 90°C for 1 h,” better quality can be expected by omitting this procedure (R) (Fig. 3).

10S LBC specimens can be stored for a long time under appropriate conditions; however, for specimens to be used for genomic testing, avoid using such specimens whenever possible (N).

FFPE Cell Block SpecimensHandling at the Time of Specimen Submission and after Specimen Processing

1. It is desirable to immerse the specimen in a formalin fixative solution for fixation after an appropriate amount of cells are collected and pretreated as needed (C).

2. When an FFPE cell block is prepared after diagnosis, it is desirable to store the remaining specimen in a refrigerator (4°C) (C) [14].

Composition of Formalin Fixative Solution

3. The desirable composition of formalin fixative solution is 10% neutral-buffered formalin (NBF) solution (C) [13].

Acceptable Duration for Formalin Fixative Solution Treatment

4. The desirable fixation time for cytological specimens is about 3–24 h, similar to that of biopsied tissues (C) [22].

5. In multilayered formalin fixation using the centrifuge tube or centrifugal cell collection method, quality deterioration caused by insufficient fixation must be avoided because deeper sediment layers tend to remain unfixed when a large amount of sedimented specimen is used (N).

Selection of FFPE Cell Block Preparation Method

6. Various preparation methods, such as the centrifugal cell collection and cell coagulation methods, are available. Therefore, it is desirable to keep two or more cell block methods ready (C) [17].

FFPE Cell Block Processing

7. Avoid effects of contamination caused by cell blocks that are difficult to solidify (N).

Evaluation of TC Contents in Cell Block Preparation

8. Cell block preparation methods that enable easy morphological evaluation of the distribution and percentage of TCs are preferred (C) [13] (Fig. 6, online suppl. Fig. 11 and 12).

9. Collect a necessary and sufficient number of nucleated cells for genomic analysis and solidify them properly (C).

10. Confirm TCs by, for example, hematoxylin and eosin (HE) staining, Papanicolaou staining, and periodic acid-Schiff reaction (C).

11. Use additional ancillary tests, such as immunocytochemical staining, to confirm TC content (C).

12. Prevent the degeneration of cells caused by sampling errors and prefixation treatment as much as possible (N).

13. Avoid using specimens stored at room temperature for more than a day (N).

Nucleic Acid Extraction from FFPE Cell Block Specimens

14. When using a commercially available DNA extraction kit for FFPE tissue, it is desirable to follow manufacturer’s instructions (C) (online suppl. Fig. 10).

15. When using a commercially available DNA extraction kit for FFPE tissue designed to carry out proteolysis with a protein K solution and removal of cross-links between nucleic acids and formalin simultaneously by incubating at a constant temperature, an appropriate condition should be used because prolonged incubation may result in the deterioration of nucleic acid quality (C) (online suppl. Fig. 10).

16. FFPE cell blocks, similar to FFPE tissues, can be stored at room temperature; however, it is desirable to avoid storing them at high temperature and humidity. They must be stored in a cool and dark place (for less than 3 years) (C) [13, 14].

Supplementary Explanations

1S In some studies, cell blocks were prepared from LBC preservative solutions without fixation using 10% NBF (R) [23, 24]; however, the effects of the storage duration on FFPE blocks prepared in such a manner have not been examined. Since 10% NBF fixation is essential in lung cancer companion diagnostics, studying the nucleic acid quality obtained from FFPE block specimens prepared from LBC preservative solutions will be necessary.

2S Refrigerated or frozen storage is known to decelerate the degradation and putrefaction caused by enzymes and microorganisms. Therefore, unless additional treatments can be performed promptly, it is recommended to store residual cytological specimens temporarily in a refrigerator to reduce the risk of DNA fragmentation (C) [14].

3−4S 10% NBF fixation is recommended for companion diagnostics (C) [13]. Since the nucleic acid quality of cancer cells is affected in a time-dependent manner even after 10% NBF fixation, attention must be paid to the fixation duration (online suppl. Fig. 9). It should also be noted that fixation with 10% NBF is completed in a length of time comparable or shorter than that required for biopsy tissue samples [22].

5S Multilayer fixation by the centrifuge tube method requires cell mass adjustment. Therefore, when there is a large quantity of sediment, prepare cell blocks after dividing the sediment into appropriate amounts in multiple test tubes.

6S Various FFPE cell block preparation methods are available in Japan and other countries [17]. The major FFPE cell block preparation methods used in Japan are the centrifugal cell collection method (centrifuge tube and collodion bag methods) and the cell solidification method (agar and sodium alginate methods), and there are multiple methods in each category. All methods have advantages and disadvantages, and available data are insufficient to compare all the methods in terms of nucleic acid quality. It is advisable to have at least two different cell block preparation methods ready since cells mixed and fixed in a formalin solution or an LBC preservative solution can be difficult to solidify depending on the specimen quantity (see online suppl. Information 1 and 2).

7S Because various cell materials are used, some specimens may be difficult to solidify depending on their conditions. When preparing FFPE cell blocks, care should be taken regarding contamination (leakage of cancer cells) in a closed automatic fixation and embedding device [25]. When preparing thin sections from FFPE blocks, due care should be taken to avoid contamination [13].

8S For body cavity fluid materials, it is recommended to evaluate the distribution and percentage of TCs in the vertical split section of cell block specimen, which is created by layering 10% NBF on the cell sediment. Each TC may be differentially distributed in the vertical section due to its specific gravity.

8−9S Caution should be exercised in evaluating the distribution and percentage of TCs because the fixation and hemolysis methods may change the cell distribution in a cell block (R).

10S HE staining is the standard method for evaluating the TC distribution; however, Papanicolaou staining may also be performed. In addition, performing the periodic acid-Schiff reaction and Alcian blue staining is recommended to evaluate cytoplasmic mucus (C).

11S For body cavity fluid materials, additional evaluation of mesothelial, epithelial, and TC markers using immunocytochemical staining, FISH, or other methods is desirable for differentiating active mesothelial cells (C).

12S A paucity of cell mass, cell collection from inappropriate sites, cell degeneration caused by methodological differences in hemolysis and fixation should be avoided as much as possible (N).

13S Genomic analysis using specimens stored at room temperature (approximately 25°C) for more than a day before fixation should be avoided whenever possible (N).

14S The protocol for QIAamp® DNA FFPE Tissue Kit (Qiagen) states to incubate at 90°C for 1 h for heat-mediated removal of cross-links between nucleic acids and formalin; however, the incubation time can be shortened to 30 min when FFPE cell blocks are used (R) (online suppl. Fig. 13). In addition, the yield and quality may be reduced if this procedure is omitted (N) (online suppl. Fig. 13).

15S When using Maxwell® FFPE Plus DNA Kit (Promega), incubation with proteinase K at 70°C for more than 4 h may result in the deterioration of nucleic acid quality (R) (online suppl. Fig. 13).

16S Long-term storage of FFPE cell blocks affects the nucleic acid quality, as is the case for FFPE tissues [24]. Therefore, storage in a cool and dark place at room temperature (approximately 25°C) is recommended whenever possible (C).

Smear SpecimensHandling of Stained Cytological Specimens

1. Institutions should have an quality control system in place and be familiarized with testing methods using stained cytological specimens (R) [13].

2. When cells are collected from stained specimens, caution should be exercised to prevent cell loss and contamination (R).

3. Avoid conducted genomic tests using stained cytological specimens stored for more than about 6 months whenever possible (N) (online suppl. Fig. 14).

4. When stained specimens are used for genetic testing, it is desirable to save cell images in the form of digital scanned/virtual slides or photomicrographs (R).

5. It is desirable to keep records of the fixation method, staining method, embedding presence/absence, storage condition, and time until nucleic acid extraction (R).

Handling of Unstained Cytological Specimens

6. Institutions should have a quality control system in place and be familiarized with testing methods (R).

7. It is desirable to confirm that the unstained specimens contain the target lesion (R).

8. When cells are collected from unstained specimens, caution should be exercised to avoid contamination (R).

9. Avoid performing genomic tests using unstained cytological specimens stored for about more than 6 months whenever possible (N).

10. It is desirable to keep records of unstained preparations, such as the storage condition and time until nucleic acid extraction (R).

Supplementary Explanations

1−5S Genetic testing performed using stained specimens is feasible; however, analyses of such tests are likely to fail more often compared to those performed using histological specimens [26, 27]. The tests are recommended to be conducted in institutions that are familiar with the tests. In addition, the results should be interpreted carefully, while taking into consideration the possibilities of obtaining false negative and false positive results. In addition, specimens stored for more than about 6 months are not suitable for certain purposes, such as panel testing, because the nucleic acid quality deteriorates during long-term storage (online suppl. Fig. 14).

Part 3: Empirical Data Related to Handling of Cytological SpecimensHandling before Specimen Processing: The Time of Temporary Storage before Specimen Processing Is Initiated Such as the Use of Puncture Needle, Rinsing Fluid, and Brush CytologyStudy Description

Specimens. Saline suspensions of cell samples were prepared by scraping cytology specimens collected from the tumor cut surface of fresh lung cancer surgical specimens (13 cases). One cell suspension sample was divided into equal portions. To evaluate storage conditions, the effects of temperature and time until nucleic acid extraction were studied.

Methods. The following conditions were compared: (1) 24 h at room temperature, (2) 3 days at room temperature, and (3) 3 days at 4°C. DNA was extracted using a commercially available extraction kit (QIAamp® DNA Mini Kit [Qiagen]) in accordance to manufacturer’s instructions. DNA Integrity Number (DIN) values were measured using a fully automated electrophoresis system (TapeStation system [Agilent]).

Study Results/Explanation of Figures

The DNA quality (DIN value) of the specimens stored in a refrigerator for 3 days decreased mildly and did not differ significantly from the DIN value of the specimens stored at room temperature for 24 h. The DIN value of the specimens stored for 3 days at room temperature decreased compared with that of the specimens stored for 24 h at room temperature; sharp decreases were observed in some cases (Fig. 2).

It is recommended to perform nucleic acid extraction as soon as possible. For long-term storage, the decrease in DNA quality can be minimized if stored in a refrigerator for no more than 3 days.

Method for Evaluating TC Contents: Analysis of Changes in the TC Content before and after the Preparation of LBC SpecimensStudy Description

Specimens. LBC specimens were prepared from suspensions containing mixtures of a lung adenocarcinoma cell line (NCI-H1975, hereafter called H1975) and a lymphoma cell line (L1236) in various ratios. Differences between the ratios of lung cancer and lymphoma cells in the LBC specimens and those in the suspensions were examined (online suppl. Fig. 1A).

Methods. The H1975 lung adenocarcinoma cell line is positive for cytokeratin 7 immunostaining, whereas the L1236 lymphoma cell line is negative (online suppl. Fig. 1B). This property allows us to distinguish H1975 cells from L1236 cells. We prepared cell suspensions (1 × 106 total cells) containing H1975 and L1236 cells in the following ratios: 100:0, 80:20, 50:50, 20:80, and 0:100. LBC specimens were prepared from each cell suspension according to a conventional method by Roche, BD, and Hologic and immunostained for cytokeratin 7. Cytokeratin 7-positive and negative cells were counted, and the ratio of these two different cells was calculated.

Study Results/Explanation of Figures

Mixtures of H1975 and L1236 cells at ratios of 100:0, 80:20, 50:50, 20:80, and 0:100 are referred to as H100%, H80%, H50%, H20%, and H0%, respectively. For each sample, cytokeratin 7 positivity rates were determined in three independent fields of view, and their mean rates are shown.

All cells in H100% were cytokeratin 7-positive, whereas all cells in H0% were cytokeratin 7-negative. The image shows cytokeratin 7 immunostaining results of the H50% mixture.

As shown in the table, the TC ratios before and after the preparation of LBC specimens were mostly similar; however, the H1975 cell contents tended to be slightly lower in Roche LBC specimens (online suppl. Fig. 1C).

The bar graph shows differences between TC ratios in the suspensions before preparation of LBC specimens and those in the LBC specimens. As mentioned above, the H1975 cell content after specimen preparation tended to be lower by up to about 4% for the Roche LBC specimens (online suppl. Fig. 1D).

Although the differences were not large, the reduction tendency mentioned above should be understood before TC contents are evaluated.

Nucleic Acid Extraction Methods: Optimal Nucleic Acid Extraction Methods for Specimens Preserved Using Different LBC Preservative SolutionsStudy Description

Specimens. Cells of a lung adenocarcinoma cell line (NCI-H1975, 6.7 × 105 cells/test) were fixed in an LBC preservative solution (CytoRichTM Red (Fig. 3a, c) and CytoRichTM Blue (for nongynecological use [BD], Fig. 3b, d), for 24 h, and the resulting sediment was used for examinations.

Methods. A commercial DNA extraction kit was used for FFPE tissues (QIAamp® DNA FFPE Tissue Kit [Qiagen] (Fig. 3a, b) and Maxwell® FFPE Plus DNA Kit [Promega] (Fig. 3c, d)). Pretreatment with proteinase K was carried out at various temperatures before DNA extraction (eluate, 50 μL). In addition, absorbance (NanoDropTM [Thermo Fisher]) and fluorescence (Qubit [Thermo Fisher]) values were measured, and a fully automated electrophoresis system (TapeStation system [Agilent]) was used to perform electrophoresis for determining nucleic acid quality.

Study Results/Explanation of Figures

Based on the absorbance measurements obtained using NanoDropTM, the concentrations were seen to increase significantly on the addition of the 56°C or 90°C pretreatment step (Fig. 3a). Similarly, concentrations that were determined fluorometrically using Qubit were seen to increase on the addition of the 56°C or 90°C pretreatment step (Fig. 3a). No major concentration differences were observed between 10 and 60-min pretreatments at 90°C (recommended).

The DIN values that indicate DNA quality was seen to increase on the addition of the 56°C or 90°C pretreatment step. DIN values tended to decrease as the duration of pretreatment performed at 90°C increased (Fig. 3a).

Study Results/Explanation of Figures

Concentrations determined based on the absorbance values measured using NanoDropTM increased when the 56°C or 90°C pretreatment step was added (Fig. 3b). The concentration increased as the duration of pretreatment performed at 90°C increased; however, no differences were seen with pretreatments performed for ≥30 min. Concentrations fluorometrically determined using Qubit did not differ depending on the duration of pretreatment (Fig. 3b).

The DIN-based comparison of DNA quality showed that DIN tended to decrease as the duration of pretreatment performed at 90°C increased (Fig. 3b).

Study Results/Explanation of Figures

Concentrations determined based on the absorbance values measured using NanoDropTM increased in a time-dependent manner for up to 2 h of pretreatment and decreased when the pretreatment was performed for 4 h (Fig. 3c). The concentrations fluorometrically determined using Qubit increased up to 1 h of pretreatment and then remained unchanged for up to 4 h (Fig. 3c).

The DIN-based comparison of DNA quality showed that DIN increased proportionally with the pretreatment duration for up to 1 h and remained unchanged for up to 4 h (Fig. 3c).

Study Results/Explanation of Figures

Concentrations determined based on the absorbance values measured using NanoDropTM tended to decrease as the duration of pretreatment increased (Fig. 3d). The concentrations fluorometrically determined using Qubit did not change significantly with the pretreatment duration (Fig. 3d).

The DIN values that indicate DNA quality did not differ with different pretreatment durations (Fig. 3d).

The optimum condition for performing proteinase K pretreatment differs depending on the fixative and extraction method used. For instance, differences in the pretreatment process have greater effects on the yield, and quality of DNA, when CytoRichTM Red, rather than CytoRichTM Blue, is used for fixation.

Caution should be exercised when interpreting measurements obtained using NanoDropTM and Qubit because they are not always similar.

LBC Specimen Preparation Methods: Changes in the DNA/RNA Yield and Quality of Cells Stored in a Frozen State for a Long Time after Treatment with LBC Preservative SolutionStudy Description

Specimens. NCI-H2228 cells treated/fixed with three different LBC preservative solutions (ThinPrep® PreservCyt, CelVerseTM, CytoRichTM Red) and 10% NBF at room temperature for 1 or 7 days were washed with PBS. The cells were then stored at −80°C for 6 months and used to evaluate changes in the DNA/RNA yield/quality after long-term frozen storage.

Methods. DNA and RNA were extracted from 1 × 105 cells (Countess® II [Thermo Fisher] was used to count cells.) using four commercially available nucleic acid extraction kits specified in the figure below. Nucleic acid concentrations were determined by measuring absorbance (NanoDropTM [Thermo Fisher]) and fluorescence (Qubit® [Thermo Fisher]), and DIN and DV200 values were measured using a fully automated electrophoresis system (TapeStation system [Agilent]) (n = 3). Papanicolaou staining was used to evaluate the effects of long-term frozen storage on cell morphology.

Study Results/Explanation of Figures

The yield and quality of DNA (Fig. 4) and RNA (Fig. 5) of cells kept under frozen storage for a long time after treatment with LBC preservative solutions were comparable with those of DNA and RNA of cells kept under frozen storage for the same period of time without LBC treatment; however, index values for the DNA/RNA yield and quality varied substantially depending on the combination of LBC preservative solution and nucleic acid extraction kit used. DNA and RNA from cells treated/fixed with CytoRichTM Red would be extracted with commercially available kits for FFPE samples but not for cultured cell samples. The quality evaluated by DIN and RIN was significantly high in DNA and RNA extracted from cells treated/fixed with CytoRichTM Red. The amount and quality of DNA and RNA were comparable between cells treated/fixed with ThinPrep® PreservCyt and with CelVerseTM.

Fig. 4.

Changes in the DNA yield and quality of cells stored in a frozen state for a long time after treatment with LBC preservative solution. The yield and quality of DNA of cells kept under frozen storage for a long time after treatment with LBC preservative solutions were comparable with those of DNA of cells kept under frozen storage for the same period of time without LBC treatment. ー, stored in a frozen state without treatment; TP, ThinPrep® PreservCyt (Hologic)-treated; CV, CelVerseTM (Sysmex)-treated; CR-R, CytoRichTM Red (BD)-treated; NBF, neutral-buffered formalin-fixed. a, n = 2; b, unmeasurable (below detection limit); c, not measured.

Fig. 5.

Changes in the RNA yield and quality of cells stored in a frozen state for a longtime after treatment with LBC preservative solution. The yield and quality of RNA of cells kept under frozen storage for a long time after treatment with LBC preservative solutions were comparable with those of RNA of cells kept without LBC treatment. ー, stored in a frozen state for the same period without treatment; TP, ThinPrep®TP; ThinPrep® PreservCyt (Hologic)-treated; CV, CelVerseTM (Sysmex)-treated; CR-R, CytoRichTM Red (BD)-treated; NBF, neutral-buffered formalin-fixed. a, n = 2; b, unmeasurable (below detection limit); c, not measured.

Observation of morphology was possible for cells kept under long-term frozen storage after LBC preservative solution treatment (online suppl. Fig. 2).

LBC Specimen Preparation Methods, Morphological Changes, and Effects on DNA Caused by Storage Using Various LBC Preservative SolutionsStudy Description

Specimens. After submitting body cavity fluid samples, clinical ovarian cancer specimens (only one case) prepared with the following LBC preservative solutions were stored for 4 days. Cell blocks were prepared by the sodium alginate method and used to confirm morphological changes and effects on DNA.

CytoRichTM Red (for nongynecological use [BD])

ThinPrep® PreservCyt (for nongynecological use [Hologic])

Cellprep® (preservative solution for fine-needle aspiration and body cavity fluid samples [Roche Diagnostics])

Methods. Cobas DNA preparation kit (FFPE) and an automated extraction device (QIAcube [Qiagen]) were used for DNA extraction. After DNA extraction, DINs values were measured using a fully automated electrophoresis system (TapeStation system [Agilent]). Morphological observation was performed using HE and immunostaining methods (WT1 antibody).

Study Results/Explanation of Figures

Practically, no morphological changes (HE staining) were observed for cell blocks prepared using the sodium alginate method and preserved using various LBC preservative solutions, indicating no significant effects on the morphology (online suppl. Fig. 3A).

DIN values for cell blocks prepared using the sodium alginate method and preserved using various LBC preservative solutions indicated relatively high DNA quality; therefore, the preservative solutions had virtually no effects on DNA quality (online suppl. Fig. 3B).

Cell blocks prepared using the sodium alginate method and stored in various LBC preservative solutions are compatible with immunostaining (online suppl. Fig. 3C).

LBC Specimen Preparation Methods, Effects of Differences in Storage Duration, and Nucleic Acid Extraction Method on DNA Quality/Stability in a Lung Adenocarcinoma Cell Line Fixed Using ThinPrep® PreservCytStudy Description

Specimens. NCI-H1975 cells (1 × 106) were fixed in 1 mL of ThinPrep® PreservCyt solution (for nongynecological use [Hologic]) and stored at room temperature (for 15 min and 1, 3, 5, 7, and 9 days). Next, genomic DNA extracted by two different methods were examined for the effects of the storage duration and extraction method on DNA stability (n = 5).

Methods. DNA extraction was carried out with an extraction kit (QIAamp® DNA Mini Kit [Qiagen]) using two different protocols (without ribonuclease [RNase] use) – one for tissues and another for cultured cells. The total nucleic acid amounts were measured using absorbance values (Spectrophotometer [DeNovix]), and the dsDNA yields and DINs were measured by an automated electrophoresis system (TapeStation system [Agilent]). Furthermore, cycleave PCR method was used to detect EGFR exon21 p.L858R point mutation, and the cycle threshold (Ct) values were compared to evaluate nucleic acid quality.

Study Results/Explanation of Figures

Total nucleic acid and double-strand DNA amounts: the total nucleic acid amount began to decrease on day 7 when the protocol for tissues was used. When the protocol for cultured cells was used, the total nucleic acid amount tended to decrease on day 9. On comparing the two protocols, the yields obtained with the protocol for tissues were observed to be higher regardless of the fixation duration, except for a 7-day fixation duration (online suppl. Fig. 4A). Conversely, there were virtually no differences in the amount of dsDNA obtained using the two protocols regardless of the fixation duration. Sharp yield decreases were observed after a 9-day fixation duration (73.7% decrease with the protocol for tissues and 78.5% decrease with the protocol for cultured cells) compared with those observed after a 15-min fixation duration (online suppl. Fig. 4B).

DINs: DINs decreased after a 1-day fixation and remained almost constant until day 9. DINs obtained with the two protocols were comparable, although the protocol for cultured cells gave slightly higher values (online suppl. Fig. 4C).

Ct values of quantitative PCR by cycleave PCR method: Ct values did not change substantially as per the fixation duration and remained sufficiently detectable even on day 9 when the dsDNA amount decreased. No major differences were found between the two protocols regardless of the fixation duration, except for the 7-day fixation duration with which the protocol for cultured cells gave a slightly lower Ct value (the smallest and largest Ct differences were 0.28 on day 9 and 0.47 on day 1, respectively) (online suppl. Fig. 4D).

LBC Specimen Preparation Methods, Effects of Differences in the Storage Period, and Nucleic Acid Extraction Method on the DNA Quality/Stability in a Lung Adenocarcinoma Cell Line Fixed with Cellprep®Study Description

Specimens. NCI-H1975 cells (1 × 106) were fixed in 1 mL of Cellprep® solution (for respiratory use [Roche Diagnostics]) and stored at room temperature (for 30 min, 1, 3, 5, 7, and 9 days). Then, genomic DNA extracted by two different methods were examined for the effects of the storage period and extraction method on the DNA stability (N = 5).

Methods. DNA extraction was carried out with an extraction kit (QIAamp® DNA Mini Kit [Qiagen]) using two different protocols (without RNase use) – one for tissues and another for cultured cells. The total nucleic acid amounts were measured using absorbance values (Spectrophotometer [DeNovix]), and the dsDNA yields and DINs were measured by an automated electrophoresis system (TapeStation system [Agilent]). Furthermore, cycleave PCR method was used to detect EGFR exon21 p. L858R point mutation, and the Ct values were compared to evaluate nucleic acid quality.

Study Results/Explanation of Figures

Total nucleic acid and dsD