Abstract
Background
The lung cancer biopsy specimens obtained by endobronchial ultrasound-guide sheath (EBUS-GS) trans lung biopsy occasionally do not contain cancer cells. It is a problem that there is a possibility that they may not contain cancer cells.
Aim of the study
To investigate the proportion of biopsy specimens containing cancer cells in total biopsy specimens.
Materials & methods
Patients with lung cancer diagnosed by EBUS-GS were selected. The primary end point was the proportion of specimens containing tumors in the total specimens obtained by EBUS-GS.
Results
Twenty-six patients were investigated. The percentage of specimens containing cancer cells in the total specimens was 79.0%.
Conclusion
The proportion of biopsy specimens containing cancer cell to all biopsy specimens by EBUS-GS was high, but not 100%.
Keywords: biopsy, EBUS-GS, lung cancer, proportion
Plain language summary
The lung cancer biopsy specimens obtained by trans lung biopsy occasionally do not contain cancer cells. It is a problem that there is a possibility that they may not contain cancer cells. If the biopsy specimens contained no malignant cells, there was a possibility that accurate genetic test would not be performed and as a result, the correct molecular targeted drug would not be used. It is important to investigate the proportion of biopsy specimens containing cancer cells in total biopsy specimens. In this study, we showed the percentage of specimens containing cancer cells in the total specimens was 79.0%. This result should be considered to perform genetic test from biopsy specimens.
Molecular profiling of a patient’s tumor to guide targeted treatment selection offers the potential to advance patient care by improving outcomes and minimizing toxicity. It sometimes find the ‘druggable’ mutations such as EGFR, the most impactful molecular alteration to date, and an excellent predictive biomarker in treating NSCLC [1]. To investigate the profile, next generation sequencing-based in vitro diagnostics tests, such as Oncomine DxTT and FoundationOne CDx panels, are used in clinical settings [2,3]. To examine the molecular profile, formalin-fixed paraffin embedded (FFPE) tissue biospecimens were usually used. However, the quality of nucleic acids is of concern [4]. Fresh frozen (FF) samples are still considered to be more suitable for molecular profiling of cancer than FFPE in terms of the quality of nucleic acids [4,5]. To investigate peripheral lung nodules, the choice of methods is usually a bronchoscopy. Endobronchial ultrasonography with a guide sheath (EBUS-GS) improves the diagnosis and is a standard method [6]. However, the biopsy specimens obtained by EBUS-GS occasionally do not contain tumor cells, and it can be difficult to confirm whether they contain tumor cells in specimens on site. When the specimens are stored as an FF, there is a possibility that they may not contain cancer cells. This is a major problem in preserving biopsy specimens as an FF sample. To overcome this problem, it might be recommended that plural specimens are preserved in a form of FF. The proportion of biopsy specimens containing cancer cells in the total biopsy specimens is critical. The higher the proportion, the fewer specimens that must be preserved. It is important to know the proportion of biopsy specimens containing tumors to all biopsy specimens using EBUS-GS for lung cancer patients.
Aim of study
To investigate the proportion of biopsy specimens containing cancer cells in total biopsy specimens.
Materials & methods
Study design
This was a single institutional retrospective observational study.
Selection of patients
Patients suspected of having lung cancer between October 2019 and March 2020 at Nagasaki University Hospital were screened. Patients with lung cancer diagnosed histologically by EBUS-GS were selected. The enrolled patients were not previously discussed in multidisciplinary teams but physician’s choice in order to define the most appropriate bioptic route.
Data collection
The following characteristics were collected from electronic medical records: age, sex, pathology, smoking status, staging, size of lesion, characteristics (pure ground grass nodule, mixed ground grass nodule, solid or cavity), position of nodule (within 1/3 from peripheral or not on chest x-ray), type of forceps (small: 2 mm, large: 2.6 mm), echo findings (none, adjacent to, within) [6], number of transbronchial biopsy (TBB) specimens and number of specimens containing cancer cells. The pathologist confirmed the number of total TBB specimens and the number of specimens containing cancer cells.
Statistical analysis
The primary end point was the proportion of specimens containing tumors in the total specimens obtained by TBB using EBUS-GS. The formula for calculating the proportion was (specimens containing tumor/total specimens) ×100. Continuous variables were analyzed using the Student’s t-test. The correlation of continuous variables was analyzed using the Pearson product–moment correlation coefficient.
Results
A total of 56 patients with suspected lung cancer were screened. The final diagnosis for 20 patients was not lung cancer. Ten cases were lung cancer, and cancer cells were not confirmed by EBUS-GS. Twenty-six patients were diagnosed with lung cancer, and this cohort was the object of analysis (Figure 1). The patient characteristics are shown in Table 1. The median age was 77.5 years old and the median size of the target lesion was 3.0 cm. The median number of TBBs was 9.
Figure 1. . Overview of patient selection.
EBUS-GS: Endobronchial ultrasound-guide sheath.
Table 1. . Patient characteristics.
| Factor | Group | Overall (n = 26), median (range)/n (%) |
|---|---|---|
| Age (years) | 77.5 (59–91) | |
| Size (cm) | 3.0 (1.3–7.0) | |
| Number of TBB | 9 (1–14) | |
| Number of years as a doctor | 4 (3–18) | |
| Echo findings | Adjacent to | 5 (19.2) |
| Within | 21 (80.8) | |
| Staging | I | 10 (38.5) |
| II | 1 (3.8) | |
| III | 7 (26.9) | |
| IV | 6 (23.1) | |
| Relapse | 2 (7.7) | |
| Lesion characteristics | Part solid | 2 (7.7) |
| Solid | 24 (92.3) | |
| Site | LLL | 8 (30.8) |
| LUL | 7 (26.9) | |
| RLL | 3 (11.5) | |
| RML | 4 (15.4) | |
| RUL | 4 (15.4) | |
| Sex | Male | 15 (57.7) |
| Female | 11 (42.3) | |
| Size of forceps | 1.5 mm | 7 (26.9) |
| 1.9 mm | 19 (73.1) | |
| Smoking | 0 | 10 (38.5) |
| 1 | 16 (61.5) | |
| Within 1/3 from peripheral | No | 8 (30.8) |
| Yes | 18 (69.2) |
LUL: Left upper lobe; LLL: Left lower lobe; RUL: Right upper lobe; RML: Right middle lobe; RLL: Right lower lobe; TBB: Transbronchial biopsy.
In this study, echo findings in EBUSGS were divided into three groups, namely ‘within’, ‘adjacent to’ and ‘invisible’. Interestingly, our results indicated that 81% of cases were found ‘within’ (n = 21). Stage I was the most common (ten cases, 38.5%), followed by stage IV (six cases, 23.1%). Solid was the most common lesion characteristic on chest computed tomography (24 cases, 92.3%). The lesion site included the ever lobe, and the left lower lobe was high (eight cases, 30.8%). Fifteen cases (57.7%) were male, and (16 cases, 61.5%) had a smoking history. Larger forceps were used in the majority of cases (19 cases, 73.1%). Peripheral lesions were the most common (18 cases, 69.2%). The percentage of specimens containing cancer cells in all specimens, which was the primary end point, was 79.0% (95% CI: 8.1–89.8) (Figure 2). The subgroup analysis is shown in Figure 3. Within the echo findings, 3 cm or more target lesion, more central position and smaller forceps were factors that tended to have a higher percentage but not statistically significant in any factors. Correlation analysis between the number of TBB was performed and did not show a significant correlation (Figure 4).
Figure 2. . Percentage of specimens containing cancer cells in the total specimens.
Figure 3. . Percentage of specimens containing cancer cells in the total specimens by subgroup.
Figure 4. . Correlation analysis between percentage of specimens containing cancer cells in the total specimens and number of transbronchial biopsy (Pearson correlation coefficient).
Discussion
To the best of our knowledge, this is the first study to investigate the proportion of specimens containing tumors in TBB by EBUS-GS. It was shown that not all TBB specimens obtained by EBUS-GS contained cancer cells.
There are several reports regarding the importance of preserving specimens as FF. Sakata et al. reported that the success rate of genetic alteration testing using Oncomine DxTT for four genes (EGFR, ALK, ROS1, BRAF) in 533 NSCLC patients was 80.1% (95% CI: 76.5–83.4%) [3]. In addition, the success rate using RNA was affected by the quality of the tumor samples rather than by the tumor cell content. In this study, all specimens used in the test were FFPE [3]. Hedegaard et al. compared 38 paired FF/FFPE specimens for NGS-based exome and transcriptome analyses. DNA Exome-Seq data showed concordance for 70–80% of variants in FF and FFPE samples, and RNA-Seq data revealed a high correlation of expression profiles in FF/FFPE pairs (Pearson correlations of 0.90 +/- 0.05) [5]. FF samples are suitable form for molecular profiling of cancer in terms of the quality of nucleic acids.
With reference to the result of this study, if one sample was stored as FF, the probability that the samples do not contain cancer cells was 21%. If three samples are stored, the probability that the samples do not contain cancer cells can be reduced to 0.9%. Clinicians sometimes avoid storing specimens as FF because they may not contain tumor. However, with reference to the result of this study, storing specimens as 3 or more FFs reduces the risk of tumors not being included. The basis for how many FFs should be preserved when performing EBUS-GS was not clear, but this study has clarified.
To investigate the factors that contribute to the proportion of specimens containing tumors, several covariates were analyzed. For echo findings, tumor sized, more peripheral position (within a third of peripheral lung) or larger forceps tended to be factors that affected the proportion of specimens containing cancer cells in the total specimens. It was expected that the diagnostic yield of EBUS-GS for peripheral lung lesions ≥3 cm, having echo findings ‘within’ or larger forceps improved over time [7]. Similarly, if 3 cm or more, it was thought that the proportion of specimens containing tumors might increase. Because the bronchial wall becomes thinner as it becomes more distal, lung cancer in more peripheral position can be easily harvested without obstruction by the bronchial wall. As expected, having echo findings ‘within’ and 3 cm or more tumor size tended to be higher proportion of biopsy specimens containing a tumor (Figure 3). Contrary to expectations, smaller forceps and more central position tended to be higher (Figure 3). It was thought that proportion of biopsy specimens containing a tumor would be high if more biopsies were taken. There was no correlation between number of TBB and the proportion of biopsy specimens containing a tumor (Figure 4). Considering these results, we need to strive for a finding of ‘within’, not ‘adjacent to’ or ‘invisible’ to improve the proportion of biopsy specimens containing tumors to all biopsy specimens by TBB using EBUS-GS.
Limitation of this study
Sample size in this study was small. This was an underpowered study to verify which factors affected the proportion of the specimens containing tumors. Single institution study was also limitations of this study. EBUS-GS is a test that requires skill and technical bias is a concern. Selection bias was also concerned because the enrolled patients were not previously discussed in multidisciplinary teams.
To address these limitations and make this study more robust, a study that includes more cases and more institutions is planned. If the proportion of specimens containing tumors was relieved, more accurate and efficient collection of cancer cells becomes possible and the quality of genetic testing for cancer treatments can be improved.
Conclusion
The proportion of biopsy specimens containing tumors to all biopsy specimens by TBB using EBUS-GS for lung cancer patients was high, but not 100%. This is potentially useful information, especially for preservation the specimens for which the presence of cancer cells cannot be directly confirmed. A study incorporating more cases is planned.
Summary points.
Molecular profiling of NSCLC patients is important to treat with molecular target drug.
Biopsy specimens, obtained by EBUS-GS occasionally do not contain tumor cells, and difficulty to confirm whether they contain tumor cells in specimens on site is a problem.
To investigate the proportion of biopsy specimens containing cancer cells in total biopsy specimens in this study.
The percentage of specimens containing cancer cells in all specimens was 79.0%.
This is the first study to investigate the proportion of specimens containing tumors in transbronchial biopsy by EBUS-GS.
Acknowledgments
English proofreading of this study was supported by Editage.
Footnotes
Author contributions
R Tagawa, S Ono, N Honda, T Suyama, Y Umeyama, Y Dotsu, H Gyotoku, H Yamaguchi, N Sakamoto, Y Obase, M Fukuda and H Mukae collected datasets. M Ozasa checked the proportion of specimens containing the tumor as a pathologist. All authors read and approved the final manuscript.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Medical writing support was provided by Editage.
Ethical conduct of research
The study protocol was reviewed and approved by the Institutional Review Board of Nagasaki University Hospital. Informed consent was obtained in the form of an opt-out on the website. Consent for publication was not applicable. This study was performed in accordance with the principles of the Declaration of Helsinki.
Data availability statement
All data generated or analyzed during this study are included in this published article.
Open access
This work is licensed under the Attribution-NonCommercial-NoDerivatives 4.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/
References
Papers of special note have been highlighted as: • of interest; •• of considerable interest
- 1.Soria J-C, Ohe Y, Vansteenkiste J et al. Osimertinib in untreated EGFR-mutated advanced non-small-cell lung cancer. N. Engl. J. Med. 378(2), 113–125 (2018). [DOI] [PubMed] [Google Scholar]; • Shows the importance of molecular testing in NSCLC.
- 2.Sun X, Zuo K, Yao Q et al. Invasive apocrine carcinoma of the breast: clinicopathologic features and comprehensive genomic profiling of 18 pure triple-negative apocrine carcinomas. Mod. Pathol. 33(12), 2473–2482 (2020). [DOI] [PubMed] [Google Scholar]; • Shows the importance of molecular testing in breast cancer. Genetic test in any carcinoma should be done.
- 3.Sakata S, Otsubo K, Yoshida H et al. Real-world data on NGS using the oncomine DxTT for detecting genetic alterations in non-small-cell lung cancer: WJOG13019L. Cancer Sci. 113(1), 221–228(2021). [DOI] [PMC free article] [PubMed] [Google Scholar]; • Shows comprehensive genetic test was done in clinical setting.
- 4.Greytak SR, Engel KB, Bass BP, Moore HM. Accuracy of molecular data generated with FFPE biospecimens: lessons from the literature. Cancer Res. 75(8), 1541–1547 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]; • Shows the accuracy of genetic test of formalin-fixed paraffin embedded or fresh frozen samples.
- 5.Hedegaard J, Thorsen K, Lund MK et al. Next-generation sequencing of RNA and DNA isolated from paired fresh-frozen and formalin-fixed paraffin-embedded samples of human cancer and normal tissue. PLOS ONE 9(5), e98187 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]; • Shows the accuracy of genetic test of formalin-fixed paraffin embedded or fresh frozen samples.
- 6.Kurimoto N, Miyazawa T, Okimasa S et al. Endobronchial ultrasonography using a guide sheath increases the ability to diagnose peripheral pulmonary lesions endoscopically. Chest 126(3), 959–965 (2004). [DOI] [PubMed] [Google Scholar]; •• Shows endobronchial ultrasound-guide sheath is a standard method to get specimens to test genetic status.
- 7.Steinfort DP, Vincent J, Heinze S, Antippa P, Irving LB. Comparative effectiveness of radial probe endobronchial ultrasound versus CT-guided needle biopsy for evaluation of peripheral pulmonary lesions: a randomized pragmatic trial. Resp Med 105(11), 1704–1711 (2011). [DOI] [PubMed] [Google Scholar]; •• Shows the diagnostic yield of endobronchial ultrasound-guide sheath for peripheral lung lesions ≥3 cm, having echo findings ‘within’ or larger forceps improved over time.