AJCP / Original Article
Adequacy of Core Needle Biopsy Specimens and Fine-Needle Aspirates for Molecular Testing of Lung Adenocarcinomas Frank Schneider, MD, Matthew A. Smith, MD, Molly C. Lane, Liron Pantanowitz, MD, Sanja Dacic, MD, PhD, and N. Paul Ohori, MD From the Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA.
Am J Clin Pathol February 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
ABSTRACT Objectives: Molecular testing of lung adenocarcinomas for epidermal growth factor (EGFR) mutations and an anaplastic lymphoma kinase (ALK) translocation is important to guide directed therapy with tyrosine kinase inhibitors. The goal of this study was to determine whether transthoracic computed tomography–guided core needle biopsy (CNB) and fine-needle aspiration (FNA) biopsy specimens were equally suitable for molecular testing. Methods: We determined the percentage of 52 CNB and 120 FNA specimens that contained sufficient paraffin-embedded tumor tissue for EGFR, KRAS, and ALK testing over a period of 2 years. We correlated sample sufficiency with the sampling method, tumor size, biopsy operator, pathologist assessing the adequacy of the sample, and the number of FNA passes performed. Results: Univariate analysis showed that CNB specimens provided a significantly higher number of samples sufficient for molecular testing than did FNA specimens (67% vs 46%; P = .007) and that one operator achieved a significantly higher percentage of sufficient FNA specimens. Binomial logistic regression found sufficiency of FNA samples to correlate with tumor size (P = .015) but not operator. Conclusions: When paraffin-embedded tissue is used for molecular testing of lung cancer, CNB specimens are more likely than FNA specimens to provide adequate tissue for molecular testing. Obtaining a sufficient FNA specimen depends on the tumor size and the individual performing the biopsy.
© American Society for Clinical Pathology
Upon completion of this activity you will be able to: • describe recommended molecular testing to select lung cancer patients for targeted therapies. • discuss advantages and disadvantages of computed tomography– guided core needle biopsy and fine-needle aspiration for sampling of lung cancer for molecular testing. • recommend sample type and cellularity optimal for molecular testing for targeted therapy in lung cancer. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 306. Exam is located at www.ascp.org/ajcpcme.
Ancillary molecular testing of lung cancer has gained importance since the advent of tyrosine kinase inhibitors as therapeutic options for lung adenocarcinoma. The presence or absence of certain driver mutations, including epidermal growth factor receptor (EGFR) mutations, anaplastic large cell lymphoma kinase (ALK) gene rearrangement, or hepatocyte growth factor receptor (MET) gene amplification, helps predict therapy response and prognosis. The College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology recently proposed guidelines for molecular testing to help select patients with lung cancer for therapy with EGFR and ALK tyrosine kinase inhibitors.1 This evidence-based consensus recommends EGFR and ALK testing in all patients with advanced-stage adenocarcinoma, regardless of sex, race, smoking history, or other clinical risk factors, and to prioritize EGFR and ALK testing over other molecular predictive tests. The rapid discovery of molecular
Am J Clin Pathol 2015;143:193-200 193 DOI: 10.1309/AJCPMY8UI7WSFSYY
CME/SAM
Key Words: Fine-needle aspiration; Core needle biopsy; Lung cancer; Adenocarcinoma; Adequacy; Molecular testing
Schneider et al / Lung Cancer Molecular Testing
abnormalities in cancer will likely lead to the identification of additional therapeutic targets in the near future. Molecular testing of lung adenocarcinomas requires procurement of adequate involved tissue. Specimens suitable for testing include formalin-fixed, paraffin-embedded (FFPE) tissue and fresh (unfixed), frozen, and alcohol-fixed tissue. When using cytology samples, cell block preparations are preferred over direct smears.1-5 Some have successfully employed for testing needle rinses directly or tumor cells taken off direct smears or liquid-based preparations.4,6-8 Recommended testing methods include polymerase chain reaction (PCR)–based assays for EGFR mutations and fluorescence in situ hybridization (FISH) assays for ALK testing.1 Surgical lung cancer resections usually offer abundant tumor tissue for testing. However, advanced lung cancers often do not undergo surgical resection. Consequently, pathologists routinely find themselves confronted with requests to perform molecular testing on small biopsy specimens obtained by computed tomography (CT)–guided core needle biopsy (CNB) or fine-needle aspiration (FNA) biopsy specimens. Although cytology samples are considered acceptable specimens for molecular testing, we found in our practice an undesirably high rate of FNA specimens that were suitable for rendering a diagnosis but insufficient for molecular testing. This prompted us to assess the adequacy for EGFR and ALK testing of CNB and FNA specimens over a period of 2 years. We report here the adequacy of each method to obtain material sufficient for molecular testing. We also discuss factors associated with successful and unsuccessful biopsies.
Materials and Methods The anatomic pathology laboratory information system of the University of Pittsburgh Medical Center was queried for all CT-guided lung FNAs and CNBs between July 2011 and June 2013 for which a diagnosis of either lung adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma, was rendered. Only CT-guided transthoracic biopsy specimens were included. Biopsy procedures were performed at different sites in our multihospital system and regional hospitals. Diagnoses were established using current guidelines.9 Rapid on-site assessments of diagnostic adequacy were performed for 118 (98%) of 120 of the FNA specimens. Molecular testing was ordered reflexively on cases with the above diagnoses. For a minority of procedures, testing had been performed previously or was not needed based on clinical circumstances. These cases were excluded from the study. Most biopsy specimens represented initial sampling of a newly found lung mass. A small minority
194 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
may have represented intrapulmonary metastases or recurrences. Two patients underwent FNA twice on different dates with separate attempts at molecular testing. Operators chose the biopsy method based on their personal preference. Six patients had CNBs and FNAs performed during the same procedure because the FNA was used for targeting the tumor or the pathologist performing the on-site assessment requested a CNB and the operator felt it was safe to obtain it. In one case, both CNB and FNA were successfully tested; in one case, neither CNB nor FNA were sufficient; in one case, molecular testing was attempted on both samples, but only the core was sufficient; and in the remaining three cases, only the CNB specimens were tested successfully and the FNA specimens excluded from the study. This study was approved by the University of Pittsburgh Institutional Review Board. CNBs were performed using 18- to 22-gauge springloaded core biopsy needles with or without a coaxial introducer needle. FNAs were performed using 22- to 25-gauge spinal or Chiba-type needles, usually without a coaxial introducer. Operators occasionally employed coaxial introducer needles for FNAs, usually in cases in which FNA was followed by CNB. Both CNB specimens and needle rinses from FNA specimens were fixed in neutral buffered 10% formalin. FNA material not used to prepare cytology smears on glass slides was spun down after fixation and the pellet placed in HistoGel (American MasterTech, Lodi, CA) before processing. After processing and paraffin-embedding using routine histology procedures, histotechnologists prepared 26 sections of 4 mm thickness. Levels 1 and 6 of FNA cell blocks and levels 1, 6, and 21 of CNB specimens were stained with H&E and sent to the attending pathologist for diagnosis. The remaining levels were used as outlined in ❚Figure 1❚. Case pathologists were responsible for assessing whether the number of tumor cells in the paraffin-embedded tissue was sufficient based on the criteria below. If borderline cases were forwarded to the laboratory, a molecular pathologist reviewed the case and made a final decision. For ALK FISH testing, a pathologist circled whole-slide recuts, and those circled areas were used for signal counting. Specimens with fewer than 100 tumor cells by manual counting were rejected. FISH was done on FFPE tumor tissues using a break-apart probe to the ALK gene (Vysis LSI ALK Dual Color, Break Apart Rearrangement Probe; Abbott Laboratories, Abbott Park, IL) per the manufacturer’s instructions. Sixty cells were scored. FISH-positive cases were defined as more than 15% split signals in tumor cells.1 For PCR-based EGFR and KRAS testing, a minimum 300 tumor cells on the H&E slide were accepted for analysis, a number that others have also found to represent a suitable sample.10 Tumor cells were manually microdissected by using a dissecting microscope. Care was taken that the
© American Society for Clinical Pathology
AJCP / Original Article
dissection target contained at least 50% tumor cells. DNA was isolated using standard laboratory procedure. For the detection of mutation, DNA was amplified with primers for exons 2 and 3 of the KRAS gene and exons 18 to 21 of the EGFR gene. Then, PCR products underwent fluorescencebased Sanger cycle sequencing in the sense and antisense directions using the BigDye Terminator version 3.1 cycle sequencing kit on the ABI 3130 (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. The sequences were analyzed using Mutation Surveyor software (SoftGenetics, State College, PA). Each case was classified as positive or negative for the EGFR and KRAS mutations based on the sequencing results. The outcome measures were (1) the percentage of FNA specimens on which molecular testing was successful and (2) the percentage of CNB specimens on which molecular testing was successful. Successful molecular testing was defined as the ability to report the presence or absence of (a) EGFR mutation, (b) KRAS mutation, and (c) ALK translocation. In response to the perceived high rate of unsatisfactory samples for molecular testing, a procedure change halfway through the study period encouraged pathologists present at an FNA procedure to request two additional needle passes solely for the cell block preparation once the biopsy specimen was felt to be adequate for a diagnosis of malignancy. We compared the primary outcomes before and after this change was implemented. The following data were collected for each procedure: person performing the biopsy, tumor size, pathologist assessing adequacy of the sample at the time of biopsy, and number of FNA passes performed and number of additional FNA passes for the cell block preparation after the sample was deemed diagnostic during the on-site assessment. Means, medians, and standard deviations were calculated using Microsoft Excel (Microsoft, Redmond, WA). Statistical associations were tested with Pearson c2 tests and binomial logistic regression using SPSS version 19 (SPSS, Chicago, IL). P value of .05 or less was considered statistically significant.
Results During the study period, 120 FNA and 52 CNB specimens were diagnosed as either lung adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma, and eligible for molecular testing. On 62 FNA (52%) and two CNB (4%) specimens, testing was not ordered upfront by the case pathologist because the number of tumor cells was insufficient. We found that 35 (67%) of 52 CNB specimens and 55 (46%) of 120 FNA specimens contained sufficient tumor for molecular testing that allowed determination of EGFR
© American Society for Clinical Pathology
Paraffin-embedded tissue sample: core biopsy or FNA cell block preparation
Level 1: H&E
Pathologist
Levels 2-5: immunohistochemistry Level 6: H&E
Pathologist
Level 7: H&E for molecular testing laboratory Levels 8-17: molecular testing including EGFR/KRAS Level 18: H&E for FISH testing laboratory Levels 19-25: FISH testing including ALK
Level 21: H&E (core biopsies only)
Pathologist
Level 26: negative control for immunohistochemistry
❚Figure 1❚ Processing of paraffin blocks of core biopsy specimens and fine-needle aspirate (FNA) cell block preparations. Multiple tissue levels are sectioned during the initial histology processing to maximize yield. FISH, fluorescence in situ hybridization.
and KRAS mutation status as well as ALK translocation status. Of the 65 inadequate FNA specimens, seven (11%) were sufficient for ALK FISH but not EGFR analysis, while another seven (11%) were sufficient for EGFR analysis but not ALK FISH. ❚Image 1❚ illustrates two examples of such FNA specimens that were insufficient for one testing modality but not the other. CNB specimens were significantly more likely to obtain sufficient material compared with FNA specimens (P = .007). Successful molecular testing of CNB specimens was not associated with tumor size or operator in a binomial logistic regression model. The mean and median number of FNA needle passes performed for both adequate and inadequate biopsy specimens are shown in ❚Table 1❚. After implementing a policy to collect two additional FNA passes for the cell block preparation, the success rate for molecular testing on FNA material increased from 45% to 48%. Although the mean number of FNA passes was higher for FNAs that allowed for successful molecular testing, we found no statistical correlation between adequacy of FNA samples for molecular testing and the number of FNA passes performed (P > .5). Eleven different operators performed FNA biopsies during the study period. The two busiest individuals performed
Am J Clin Pathol 2015;143:193-200 195 DOI: 10.1309/AJCPMY8UI7WSFSYY
Schneider et al / Lung Cancer Molecular Testing
A
B
C
D
E
F
❚Image 1❚ Cell block preparations from two fine-needle aspirates that were insufficient either for EGFR/KRAS testing (A-C) or ALK testing (D-F). Low-power views highlight the paucity of tumor cell groups (arrows). A, D, Original sections evaluated by the pathologist for establishing a diagnosis. B, Recut section containing insufficient material for EGFR and KRAS analysis. C, Successful fluorescence in situ hybridization (FISH) testing for ALK. E, Recut section for successful EGFR/KRAS testing. F, Recut section showing insufficient material for ALK FISH. (A, B, H&E, ×40; C, in situ hybridization for ALK as described in the Materials and Methods, ×400; D, E, H&E, ×100; F, H&E, ×200). 196 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
© American Society for Clinical Pathology
AJCP / Original Article
❚Table 1❚ Comparison of the Success Rate and the Number of FNA Passes Performed Before and After Obtaining Two Additional Passes for Cell Block Preparation
FNA Passes
Molecular Testing Successful, %
Before obtaining two additional FNA passes 45 After obtaining two additional FNA passes 48
No. of FNA Passes When FNA Was Sufficient for Molecular Testing
No. of FNA Passes When FNA Was Insufficient for Molecular Testing
Mean ± SD
Median (Range)
Mean ± SD
Median (Range)
3.2 ± 1.1 4.0 ± 1.3
3 (2-7) 4 (2-7)
3.5 ± 1.7 3.7 ± 1.3
3 (1-10) 3 (2-7)
FNA, fine-needle aspiration.
47 and 42 FNAs, respectively; five others performed more than one. Performance of FNA operators with regard to obtaining aspirates sufficient for molecular testing is shown in ❚Table 2❚. When comparing different operators performing FNAs, one operator was significantly associated with obtaining sufficient FNA samples in univariate analysis (P = .017). However, multivariate analysis of FNAs showed successful molecular testing to be associated with tumor size (P = .015) but not the operator (P = .61). The regression line of the multivariate model suggests that in our setting, FNA specimens of lung tumors larger than 3.5 cm had a greater than 50% chance of containing sufficient material for molecular testing ❚Figure 2❚. There was no association between successful molecular testing and individual pathologists performing adequacy assessments of the sample at the time of biopsy.
Discussion Molecular testing of adenocarcinomas is important to identify driver mutations of lung cancer and enable oncologists to select proper treatment. Our experience suggests that CT-guided CNBs are superior to CT-guided FNAs when molecular testing of adenocarcinoma is desired. Several studies have addressed the adequacy of small biopsy specimens for molecular testing, yet to our knowledge, none to date have compared the adequacy of CNBs and FNAs for EGFR and ALK testing in the same environment. Ferretti et al11 report success rates for CNBs of 72% before and 92% after the 2011 guideline published jointly by the American Thoracic Society, the European Respiratory Society, and the International Association for the Study of Lung Cancer. They performed both EGFR and ALK testing, and it is unclear what changed between the two time periods. Arcila et al12 were able to perform EGFR sequencing on 89% and 79% on their CNB and FNA specimens, respectively. Even if we included our seven samples for which EGFR but not ALK testing worked (success rate of 52%), their yield is still much higher despite their sampling of treatment-resistant cancers in which one might expect a lower success rate due to treatment-related changes. Solomon et al13 report that 89% of their CNB specimens were
© American Society for Clinical Pathology
adequate for EGFR and KRAS sequencing. In all of these studies, it is unclear whether the denominator of the success rate comprises all adenocarcinomas eligible for testing or only the cases on which testing was attempted. Cai et al,14 for example, report a 91% success rate for EGFR and ALK testing on cytology specimens but specifically include only those cases on which such testing was requested. When all cases are considered, such as by Hsiao et al,15 who obtained EGFR results on 132 (80%) of a total of 164 CNB specimens, success rates are lower and similar to ours. More in keeping with our data is the experience by Knoepp and Roh,7 who report that 37% of FNA cell block preparations were acellular and another 20% sparse or borderline. Khan et al16 found that 25% of their FNA specimens did not have enough cells for immunohistochemical stains, likely an indicator that at least as many cases would also be insufficient for molecular studies. Each practice has to decide whether reflex testing on small biopsy samples is desirable or useful. Here, this practice pattern allowed us to establish success rates of CNBs and FNAs. In our study, the mean number of FNA passes for sufficient biopsy specimens was higher than that for insufficient FNA specimens, although this difference was not statistically significant. Intuitively, a higher number of FNA passes should be more likely to yield a sample suitable for molecular testing. However, our results and published data show that an increased number of passes does not necessarily result in a higher adequacy rate.7 Instead, performing immediate on-site assessments may be the best way to ensure adequacy of the sample.17 Counting the number of biopsy passes before and after implementing our new policy to request two additional FNA passes solely for cell block preparation helped assess implementation. Although the mean number of passes did not increase by two, a tendency to obtain more tissue was apparent. Our biopsy operators do not employ coaxial biopsy needle systems for FNA specimens, so there may be a reluctance to perform two additional punctures for fear of complications once diagnostic material has been obtained. Some of this concern may be alleviated by data showing that neither the number of needle passes nor the dwell time of a coaxial needle are associated with an increased incidence of pneumothorax.18,19 We also observed a reluctance of operators to perform CNBs, usually due to
Am J Clin Pathol 2015;143:193-200 197 DOI: 10.1309/AJCPMY8UI7WSFSYY
❚Table 2❚ Adequacy of FNAs for Molecular Testing by Operator Performing the Biopsy Operator (No. of FNAs Performed)
% of FNAs With Sufficient Material for Molecular Testing
P Valuea
1 (1) 2 (1) 3 (47) 4 (3) 5 (1) 6 (9) 7 (2) 8 (11) 9 (1) 10 (2) 11 (42)
0 0 63 3 100 11 0 54 100 0 38
NS NS .017 NS NS NS NS NS NS NS NS
FNA, fine-needle aspiration; NS, not significant. a Wald test applied to binomial logistic regression.
the risk of creating a pneumothorax.20 The true risk of pneumothorax may be difficult to estimate for operators, and use of smaller core needles can reduce that risk.21,22 Our comparison shows that there are significant differences between operators and/or biopsy technique. With some FNA operators, patients had a greater than 50% chance of requiring a repeat biopsy if molecular testing was needed. Published data on operator variability in image-guided biopsies are essentially nonexistent, although they are likely subject to confidential departmental quality assurance. Our numbers have to be interpreted with caution since operator performance is influenced by numerous variables, including patient characteristics, location and depth of the targeted lesion, fidelity and imaging speed of the CT scanner, and diameter and tip design of the biopsy needles. Most important, our comparison shows that even the most skilled FNA operator did not reach the adequacy rate of CNBs. We acknowledge several limitations of this study. Our institution prefers to perform molecular testing on cytology specimens through a similar platform used for our surgical pathology samples. Therefore, we are only comparing cell block preparations from FNA material with paraffinembedded CNB specimens. Cell block preparations can be cell poor despite adequately cellular aspirates, and higher success rates could possibly be achieved by scraping tumor cells directly from cytology smears.4,7 However, we prefer to leave cytology slides intact, especially if tumor cells are scant. Pathology practices submitting their tissue samples to referral laboratories for testing may also be less inclined to give up their original diagnostic cytology smears. Introducing ALK immunohistochemistry as a surrogate marker for ALK rearrangement could possibly lower the rate of insufficient ALK FISH samples.23 We also did not control for patient age, biopsy risk factors, or location of the targeted mass. We suspect that such variations were evenly distributed and differences negligible in our study
198 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
Probability of a Sufficient FNA Sample
Schneider et al / Lung Cancer Molecular Testing
1.0 0.8 0.6 0.4 0.2 0.0 0.0
2.0
4.0 6.0 Tumor Size (cm)
8.0
10.0
❚Figure 2❚ Regression curve for the binomial logistic model shows fine-needle aspirations (FNAs) more likely to yield sufficient samples for molecular testing if tumors are larger. Logistic function: Probability of a sufficient FNA sample = 1 – {1/[1 + exp(–1.342 + (0.376 × tumor size))]}.
population that included consecutive biopsy specimens diagnosed as adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma. However, we cannot exclude that patients who underwent CNBs were “easier to biopsy.” We suspect that intratumor variability of oncogenic mutations may not be very common as initial reports suggested. Recent studies confirmed homogeneity of driver mutations in primary and metastatic lung carcinomas. For example, Yatabe et al24 suggested the concept of pseudoheterogeneity of EGFR mutations. They found identical EGFR driver mutations in multiple areas of the same tumor. However, the presence of EGFR amplification and/or contamination by normal cells may interfere with the interpretation of PCR mutation assays.24 In our setting, the main issue would be the detection limit when mutated tumor cells are admixed with nonmutated tumor cells. Taniguchi et al25 showed that EGFR mutations can still be detected in areas that contain nonmutated cells. Vignot et al26 were able to identify the driver mutation in lung cancer even if a tumor had accumulated other molecular alterations. Different pathologists could potentially influence the success rate of FNAs by not requesting testing on cell blocks that contain sufficient material. This seemed unlikely because all pathologists were educated on the minimum number of cells required by the laboratory. Also, since pathologists are aware of the importance of molecular testing, they tend to err on the side of caution and are more likely to request testing on cell blocks or CNB specimens that later turn out to be insufficient. Even if our minimum number of cells were too high, it should not influence the direct comparison between CNB and FNA specimens in this study.
© American Society for Clinical Pathology
AJCP / Original Article
The present study is unique because during the study period, we performed reflex molecular testing on eligible patients with lung carcinoma. Therefore, we can use cases that were eligible for testing but did not contain sufficient material for calculation of our adequacy rate. Most studies reporting high success rates of molecular testing do not consider in their calculation of success rates those biopsy specimens on which testing was not even attempted. Furthermore, to our knowledge, this is the only study to date that compares directly the yield of CNB and FNA specimens for EGFR, KRAS, and ALK testing in the same clinical environment. Although the 67% success rate of CNBs still leaves room for improvement, not being able to offer patients molecular testing on more than half of the FNA specimens is undesirable. For clinical environments in which only paraffin-embedded tissue is used for molecular testing of lung cancer, our data suggest that (1) CNBs are preferable over FNAs to obtain adequate tissue for molecular testing of lung adenocarcinomas, and (2) FNAs are more likely to yield sufficient samples from nodules above a certain size. Address reprint requests to Dr Schneider: Dept of Pathology, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213; [email protected].
References 1. Lindeman NI, Cagle PT, Beasley MB, et al. Molecular testing guideline for selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: guideline from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology. J Thorac Oncol. 2013;8:823-859. 2. Rekhtman N, Brandt SM, Sigel CS, et al. Suitability of thoracic cytology for new therapeutic paradigms in non-small cell lung carcinoma: high accuracy of tumor subtyping and feasibility of EGFR and KRAS molecular testing. J Thorac Oncol. 2011;6:451-458. 3. Boldrini L, Gisfredi S, Ursino S, et al. Mutational analysis in cytological specimens of advanced lung adenocarcinoma: a sensitive method for molecular diagnosis. J Thorac Oncol. 2007;2:1086-1090. 4. Billah S, Stewart J, Staerkel G, et al. EGFR and KRAS mutations in lung carcinoma: molecular testing by using cytology specimens. Cancer Cytopathol. 2011;119:111-117. 5. Smouse JH, Cibas ES, Janne PA, et al. EGFR mutations are detected comparably in cytologic and surgical pathology specimens of nonsmall cell lung cancer. Cancer. 2009;117:67-72. 6. Otani H, Toyooka S, Soh J, et al. Detection of EGFR gene mutations using the wash fluid of CT-guided biopsy needle in NSCLC patients. J Thorac Oncol. 2008;3:472-476. 7. Knoepp SM, Roh MH. Ancillary techniques on direct-smear aspirate slides: a significant evolution for cytopathology techniques. Cancer Cytopathol. 2013;121:120-128. 8. da Cunha Santos G, Liu N, Tsao MS, et al. Detection of EGFR and KRAS mutations in fine-needle aspirates stored on Whatman FTA cards: is this the tool for biobanking cytological samples in the molecular era? Cancer Cytopathol. 2010;118:450-456.
© American Society for Clinical Pathology
9. Travis W, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011;6:244-285. 10. Bozzetti C, Negri FV, Azzoni C, et al. Epidermal growth factor receptor and Kras gene expression: reliability of mutational analysis on cytological samples. Diagnostic Cytopathol. 2013;41:595-598. 11. Ferretti GR, Busser B, de Fraipont F, et al. Adequacy of CT-guided biopsies with histomolecular subtyping of pulmonary adenocarcinomas: influence of ATS/ERS/IASLC guidelines. Lung Cancer. 2013;82:69-75. 12. Arcila ME, Oxnard GR, Nafa K, et al. Rebiopsy of lung cancer patients with acquired resistance to EGFR inhibitors and enhanced detection of the T790M mutation using a locked nucleic acid–based assay. Clin Cancer Res. 2011;17:1169-1180. 13. Solomon SB, Zakowski MF, Pao W, et al. Core needle lung biopsy specimens: adequacy for EGFR and KRAS mutational analysis. AJR Am J Roentgenol. 2010;194:266-269. 14. Cai G, Wong R, Chhieng D, et al. Identification of EGFR mutation, KRAS mutation, and ALK gene rearrangement in cytological specimens of primary and metastatic lung adenocarcinoma. Cancer Cytopathol. 2013;121:500-507. 15. Hsiao SH, Chung CL, Lee CM, et al. Suitability of computed tomography–guided biopsy specimens for subtyping and genotyping of non–small-cell lung cancer. Clin Lung Cancer. 2013;14:719-725. 16. Khan KA, Zaidi S, Swan N, et al. The use of computerised tomography guided percutaneous fine needle aspiration in the evaluation of solitary pulmonary nodules. Ir Med J. 2012;105:50-52. 17. Santambrogio L, Nosotti M, Bellaviti N, et al. CT-guided fine-needle aspiration cytology of solitary pulmonary nodules: a prospective, randomized study of immediate cytologic evaluation. Chest. 1997;112:423-425. 18. Halloush RA, Khasawneh FA, Saleh HA, et al. Fine needle aspiration cytology of lung lesions: a clinicopathological and cytopathological review of 150 cases with emphasis on the relation between the number of passes and the incidence of pneumothorax. Cytopathology. 2007;18:44-51. 19. Ko JP, Shepard JO, Drucker EA, et al. Factors influencing pneumothorax rate at lung biopsy: are dwell time and angle of pleural puncture contributing factors? Radiology. 2001;218:491-496. 20. Yeow KM, Su IH, Pan KT, et al. Risk factors of pneumothorax and bleeding: multivariate analysis of 660 CT-guided coaxial cutting needle lung biopsies. Chest. 2004;126:748-754. 21. Chakrabarti B, Earis JE, Pandey R, et al. Risk assessment of pneumothorax and pulmonary haemorrhage complicating percutaneous co-axial cutting needle lung biopsy. Respir Med. 2009;103:449-455. 22. Geraghty PR, Kee ST, McFarlane G, et al. CT-guided transthoracic needle aspiration biopsy of pulmonary nodules: needle size and pneumothorax rate. Radiology. 2003;229:475-481. 23. Yi ES, Chung JH, Kulig K, et al. Detection of anaplastic lymphoma kinase (ALK) gene rearrangement in non–small cell lung cancer and related issues in ALK inhibitor therapy: a literature review. Mol Diagn Ther. 2012;16:143-150. 24. Yatabe Y, Matsuo K, Mitsudomi T. Heterogeneous distribution of EGFR mutations is extremely rare in lung adenocarcinoma. J Clin Oncol. 2011;29:2972-2977.
Am J Clin Pathol 2015;143:193-200 199 DOI: 10.1309/AJCPMY8UI7WSFSYY
Schneider et al / Lung Cancer Molecular Testing
25. Taniguchi K, Okami J, Kodama K, et al. Intratumor heterogeneity of epidermal growth factor receptor mutations in lung cancer and its correlation to the response to gefitinib. Cancer Sci. 2008;99:929-935.
200 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
26. Vignot S, Frampton GM, Soria JC, et al. Next-generation sequencing reveals high concordance of recurrent somatic alterations between primary tumor and metastases from patients with non–small-cell lung cancer. J Clin Oncol. 2013;31:2167-2172.
© American Society for Clinical Pathology
Adequacy of Core Needle Biopsy Specimens and Fine-Needle Aspirates for Molecular Testing of Lung Adenocarcinomas Frank Schneider, MD, Matthew A. Smith, MD, Molly C. Lane, Liron Pantanowitz, MD, Sanja Dacic, MD, PhD, and N. Paul Ohori, MD From the Department of Pathology, University of Pittsburgh Medical Center, Pittsburgh, PA.
Am J Clin Pathol February 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
ABSTRACT Objectives: Molecular testing of lung adenocarcinomas for epidermal growth factor (EGFR) mutations and an anaplastic lymphoma kinase (ALK) translocation is important to guide directed therapy with tyrosine kinase inhibitors. The goal of this study was to determine whether transthoracic computed tomography–guided core needle biopsy (CNB) and fine-needle aspiration (FNA) biopsy specimens were equally suitable for molecular testing. Methods: We determined the percentage of 52 CNB and 120 FNA specimens that contained sufficient paraffin-embedded tumor tissue for EGFR, KRAS, and ALK testing over a period of 2 years. We correlated sample sufficiency with the sampling method, tumor size, biopsy operator, pathologist assessing the adequacy of the sample, and the number of FNA passes performed. Results: Univariate analysis showed that CNB specimens provided a significantly higher number of samples sufficient for molecular testing than did FNA specimens (67% vs 46%; P = .007) and that one operator achieved a significantly higher percentage of sufficient FNA specimens. Binomial logistic regression found sufficiency of FNA samples to correlate with tumor size (P = .015) but not operator. Conclusions: When paraffin-embedded tissue is used for molecular testing of lung cancer, CNB specimens are more likely than FNA specimens to provide adequate tissue for molecular testing. Obtaining a sufficient FNA specimen depends on the tumor size and the individual performing the biopsy.
© American Society for Clinical Pathology
Upon completion of this activity you will be able to: • describe recommended molecular testing to select lung cancer patients for targeted therapies. • discuss advantages and disadvantages of computed tomography– guided core needle biopsy and fine-needle aspiration for sampling of lung cancer for molecular testing. • recommend sample type and cellularity optimal for molecular testing for targeted therapy in lung cancer. The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module. The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose. Questions appear on p 306. Exam is located at www.ascp.org/ajcpcme.
Ancillary molecular testing of lung cancer has gained importance since the advent of tyrosine kinase inhibitors as therapeutic options for lung adenocarcinoma. The presence or absence of certain driver mutations, including epidermal growth factor receptor (EGFR) mutations, anaplastic large cell lymphoma kinase (ALK) gene rearrangement, or hepatocyte growth factor receptor (MET) gene amplification, helps predict therapy response and prognosis. The College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology recently proposed guidelines for molecular testing to help select patients with lung cancer for therapy with EGFR and ALK tyrosine kinase inhibitors.1 This evidence-based consensus recommends EGFR and ALK testing in all patients with advanced-stage adenocarcinoma, regardless of sex, race, smoking history, or other clinical risk factors, and to prioritize EGFR and ALK testing over other molecular predictive tests. The rapid discovery of molecular
Am J Clin Pathol 2015;143:193-200 193 DOI: 10.1309/AJCPMY8UI7WSFSYY
CME/SAM
Key Words: Fine-needle aspiration; Core needle biopsy; Lung cancer; Adenocarcinoma; Adequacy; Molecular testing
Schneider et al / Lung Cancer Molecular Testing
abnormalities in cancer will likely lead to the identification of additional therapeutic targets in the near future. Molecular testing of lung adenocarcinomas requires procurement of adequate involved tissue. Specimens suitable for testing include formalin-fixed, paraffin-embedded (FFPE) tissue and fresh (unfixed), frozen, and alcohol-fixed tissue. When using cytology samples, cell block preparations are preferred over direct smears.1-5 Some have successfully employed for testing needle rinses directly or tumor cells taken off direct smears or liquid-based preparations.4,6-8 Recommended testing methods include polymerase chain reaction (PCR)–based assays for EGFR mutations and fluorescence in situ hybridization (FISH) assays for ALK testing.1 Surgical lung cancer resections usually offer abundant tumor tissue for testing. However, advanced lung cancers often do not undergo surgical resection. Consequently, pathologists routinely find themselves confronted with requests to perform molecular testing on small biopsy specimens obtained by computed tomography (CT)–guided core needle biopsy (CNB) or fine-needle aspiration (FNA) biopsy specimens. Although cytology samples are considered acceptable specimens for molecular testing, we found in our practice an undesirably high rate of FNA specimens that were suitable for rendering a diagnosis but insufficient for molecular testing. This prompted us to assess the adequacy for EGFR and ALK testing of CNB and FNA specimens over a period of 2 years. We report here the adequacy of each method to obtain material sufficient for molecular testing. We also discuss factors associated with successful and unsuccessful biopsies.
Materials and Methods The anatomic pathology laboratory information system of the University of Pittsburgh Medical Center was queried for all CT-guided lung FNAs and CNBs between July 2011 and June 2013 for which a diagnosis of either lung adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma, was rendered. Only CT-guided transthoracic biopsy specimens were included. Biopsy procedures were performed at different sites in our multihospital system and regional hospitals. Diagnoses were established using current guidelines.9 Rapid on-site assessments of diagnostic adequacy were performed for 118 (98%) of 120 of the FNA specimens. Molecular testing was ordered reflexively on cases with the above diagnoses. For a minority of procedures, testing had been performed previously or was not needed based on clinical circumstances. These cases were excluded from the study. Most biopsy specimens represented initial sampling of a newly found lung mass. A small minority
194 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
may have represented intrapulmonary metastases or recurrences. Two patients underwent FNA twice on different dates with separate attempts at molecular testing. Operators chose the biopsy method based on their personal preference. Six patients had CNBs and FNAs performed during the same procedure because the FNA was used for targeting the tumor or the pathologist performing the on-site assessment requested a CNB and the operator felt it was safe to obtain it. In one case, both CNB and FNA were successfully tested; in one case, neither CNB nor FNA were sufficient; in one case, molecular testing was attempted on both samples, but only the core was sufficient; and in the remaining three cases, only the CNB specimens were tested successfully and the FNA specimens excluded from the study. This study was approved by the University of Pittsburgh Institutional Review Board. CNBs were performed using 18- to 22-gauge springloaded core biopsy needles with or without a coaxial introducer needle. FNAs were performed using 22- to 25-gauge spinal or Chiba-type needles, usually without a coaxial introducer. Operators occasionally employed coaxial introducer needles for FNAs, usually in cases in which FNA was followed by CNB. Both CNB specimens and needle rinses from FNA specimens were fixed in neutral buffered 10% formalin. FNA material not used to prepare cytology smears on glass slides was spun down after fixation and the pellet placed in HistoGel (American MasterTech, Lodi, CA) before processing. After processing and paraffin-embedding using routine histology procedures, histotechnologists prepared 26 sections of 4 mm thickness. Levels 1 and 6 of FNA cell blocks and levels 1, 6, and 21 of CNB specimens were stained with H&E and sent to the attending pathologist for diagnosis. The remaining levels were used as outlined in ❚Figure 1❚. Case pathologists were responsible for assessing whether the number of tumor cells in the paraffin-embedded tissue was sufficient based on the criteria below. If borderline cases were forwarded to the laboratory, a molecular pathologist reviewed the case and made a final decision. For ALK FISH testing, a pathologist circled whole-slide recuts, and those circled areas were used for signal counting. Specimens with fewer than 100 tumor cells by manual counting were rejected. FISH was done on FFPE tumor tissues using a break-apart probe to the ALK gene (Vysis LSI ALK Dual Color, Break Apart Rearrangement Probe; Abbott Laboratories, Abbott Park, IL) per the manufacturer’s instructions. Sixty cells were scored. FISH-positive cases were defined as more than 15% split signals in tumor cells.1 For PCR-based EGFR and KRAS testing, a minimum 300 tumor cells on the H&E slide were accepted for analysis, a number that others have also found to represent a suitable sample.10 Tumor cells were manually microdissected by using a dissecting microscope. Care was taken that the
© American Society for Clinical Pathology
AJCP / Original Article
dissection target contained at least 50% tumor cells. DNA was isolated using standard laboratory procedure. For the detection of mutation, DNA was amplified with primers for exons 2 and 3 of the KRAS gene and exons 18 to 21 of the EGFR gene. Then, PCR products underwent fluorescencebased Sanger cycle sequencing in the sense and antisense directions using the BigDye Terminator version 3.1 cycle sequencing kit on the ABI 3130 (Applied Biosystems, Foster City, CA) according to the manufacturer’s instructions. The sequences were analyzed using Mutation Surveyor software (SoftGenetics, State College, PA). Each case was classified as positive or negative for the EGFR and KRAS mutations based on the sequencing results. The outcome measures were (1) the percentage of FNA specimens on which molecular testing was successful and (2) the percentage of CNB specimens on which molecular testing was successful. Successful molecular testing was defined as the ability to report the presence or absence of (a) EGFR mutation, (b) KRAS mutation, and (c) ALK translocation. In response to the perceived high rate of unsatisfactory samples for molecular testing, a procedure change halfway through the study period encouraged pathologists present at an FNA procedure to request two additional needle passes solely for the cell block preparation once the biopsy specimen was felt to be adequate for a diagnosis of malignancy. We compared the primary outcomes before and after this change was implemented. The following data were collected for each procedure: person performing the biopsy, tumor size, pathologist assessing adequacy of the sample at the time of biopsy, and number of FNA passes performed and number of additional FNA passes for the cell block preparation after the sample was deemed diagnostic during the on-site assessment. Means, medians, and standard deviations were calculated using Microsoft Excel (Microsoft, Redmond, WA). Statistical associations were tested with Pearson c2 tests and binomial logistic regression using SPSS version 19 (SPSS, Chicago, IL). P value of .05 or less was considered statistically significant.
Results During the study period, 120 FNA and 52 CNB specimens were diagnosed as either lung adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma, and eligible for molecular testing. On 62 FNA (52%) and two CNB (4%) specimens, testing was not ordered upfront by the case pathologist because the number of tumor cells was insufficient. We found that 35 (67%) of 52 CNB specimens and 55 (46%) of 120 FNA specimens contained sufficient tumor for molecular testing that allowed determination of EGFR
© American Society for Clinical Pathology
Paraffin-embedded tissue sample: core biopsy or FNA cell block preparation
Level 1: H&E
Pathologist
Levels 2-5: immunohistochemistry Level 6: H&E
Pathologist
Level 7: H&E for molecular testing laboratory Levels 8-17: molecular testing including EGFR/KRAS Level 18: H&E for FISH testing laboratory Levels 19-25: FISH testing including ALK
Level 21: H&E (core biopsies only)
Pathologist
Level 26: negative control for immunohistochemistry
❚Figure 1❚ Processing of paraffin blocks of core biopsy specimens and fine-needle aspirate (FNA) cell block preparations. Multiple tissue levels are sectioned during the initial histology processing to maximize yield. FISH, fluorescence in situ hybridization.
and KRAS mutation status as well as ALK translocation status. Of the 65 inadequate FNA specimens, seven (11%) were sufficient for ALK FISH but not EGFR analysis, while another seven (11%) were sufficient for EGFR analysis but not ALK FISH. ❚Image 1❚ illustrates two examples of such FNA specimens that were insufficient for one testing modality but not the other. CNB specimens were significantly more likely to obtain sufficient material compared with FNA specimens (P = .007). Successful molecular testing of CNB specimens was not associated with tumor size or operator in a binomial logistic regression model. The mean and median number of FNA needle passes performed for both adequate and inadequate biopsy specimens are shown in ❚Table 1❚. After implementing a policy to collect two additional FNA passes for the cell block preparation, the success rate for molecular testing on FNA material increased from 45% to 48%. Although the mean number of FNA passes was higher for FNAs that allowed for successful molecular testing, we found no statistical correlation between adequacy of FNA samples for molecular testing and the number of FNA passes performed (P > .5). Eleven different operators performed FNA biopsies during the study period. The two busiest individuals performed
Am J Clin Pathol 2015;143:193-200 195 DOI: 10.1309/AJCPMY8UI7WSFSYY
Schneider et al / Lung Cancer Molecular Testing
A
B
C
D
E
F
❚Image 1❚ Cell block preparations from two fine-needle aspirates that were insufficient either for EGFR/KRAS testing (A-C) or ALK testing (D-F). Low-power views highlight the paucity of tumor cell groups (arrows). A, D, Original sections evaluated by the pathologist for establishing a diagnosis. B, Recut section containing insufficient material for EGFR and KRAS analysis. C, Successful fluorescence in situ hybridization (FISH) testing for ALK. E, Recut section for successful EGFR/KRAS testing. F, Recut section showing insufficient material for ALK FISH. (A, B, H&E, ×40; C, in situ hybridization for ALK as described in the Materials and Methods, ×400; D, E, H&E, ×100; F, H&E, ×200). 196 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
© American Society for Clinical Pathology
AJCP / Original Article
❚Table 1❚ Comparison of the Success Rate and the Number of FNA Passes Performed Before and After Obtaining Two Additional Passes for Cell Block Preparation
FNA Passes
Molecular Testing Successful, %
Before obtaining two additional FNA passes 45 After obtaining two additional FNA passes 48
No. of FNA Passes When FNA Was Sufficient for Molecular Testing
No. of FNA Passes When FNA Was Insufficient for Molecular Testing
Mean ± SD
Median (Range)
Mean ± SD
Median (Range)
3.2 ± 1.1 4.0 ± 1.3
3 (2-7) 4 (2-7)
3.5 ± 1.7 3.7 ± 1.3
3 (1-10) 3 (2-7)
FNA, fine-needle aspiration.
47 and 42 FNAs, respectively; five others performed more than one. Performance of FNA operators with regard to obtaining aspirates sufficient for molecular testing is shown in ❚Table 2❚. When comparing different operators performing FNAs, one operator was significantly associated with obtaining sufficient FNA samples in univariate analysis (P = .017). However, multivariate analysis of FNAs showed successful molecular testing to be associated with tumor size (P = .015) but not the operator (P = .61). The regression line of the multivariate model suggests that in our setting, FNA specimens of lung tumors larger than 3.5 cm had a greater than 50% chance of containing sufficient material for molecular testing ❚Figure 2❚. There was no association between successful molecular testing and individual pathologists performing adequacy assessments of the sample at the time of biopsy.
Discussion Molecular testing of adenocarcinomas is important to identify driver mutations of lung cancer and enable oncologists to select proper treatment. Our experience suggests that CT-guided CNBs are superior to CT-guided FNAs when molecular testing of adenocarcinoma is desired. Several studies have addressed the adequacy of small biopsy specimens for molecular testing, yet to our knowledge, none to date have compared the adequacy of CNBs and FNAs for EGFR and ALK testing in the same environment. Ferretti et al11 report success rates for CNBs of 72% before and 92% after the 2011 guideline published jointly by the American Thoracic Society, the European Respiratory Society, and the International Association for the Study of Lung Cancer. They performed both EGFR and ALK testing, and it is unclear what changed between the two time periods. Arcila et al12 were able to perform EGFR sequencing on 89% and 79% on their CNB and FNA specimens, respectively. Even if we included our seven samples for which EGFR but not ALK testing worked (success rate of 52%), their yield is still much higher despite their sampling of treatment-resistant cancers in which one might expect a lower success rate due to treatment-related changes. Solomon et al13 report that 89% of their CNB specimens were
© American Society for Clinical Pathology
adequate for EGFR and KRAS sequencing. In all of these studies, it is unclear whether the denominator of the success rate comprises all adenocarcinomas eligible for testing or only the cases on which testing was attempted. Cai et al,14 for example, report a 91% success rate for EGFR and ALK testing on cytology specimens but specifically include only those cases on which such testing was requested. When all cases are considered, such as by Hsiao et al,15 who obtained EGFR results on 132 (80%) of a total of 164 CNB specimens, success rates are lower and similar to ours. More in keeping with our data is the experience by Knoepp and Roh,7 who report that 37% of FNA cell block preparations were acellular and another 20% sparse or borderline. Khan et al16 found that 25% of their FNA specimens did not have enough cells for immunohistochemical stains, likely an indicator that at least as many cases would also be insufficient for molecular studies. Each practice has to decide whether reflex testing on small biopsy samples is desirable or useful. Here, this practice pattern allowed us to establish success rates of CNBs and FNAs. In our study, the mean number of FNA passes for sufficient biopsy specimens was higher than that for insufficient FNA specimens, although this difference was not statistically significant. Intuitively, a higher number of FNA passes should be more likely to yield a sample suitable for molecular testing. However, our results and published data show that an increased number of passes does not necessarily result in a higher adequacy rate.7 Instead, performing immediate on-site assessments may be the best way to ensure adequacy of the sample.17 Counting the number of biopsy passes before and after implementing our new policy to request two additional FNA passes solely for cell block preparation helped assess implementation. Although the mean number of passes did not increase by two, a tendency to obtain more tissue was apparent. Our biopsy operators do not employ coaxial biopsy needle systems for FNA specimens, so there may be a reluctance to perform two additional punctures for fear of complications once diagnostic material has been obtained. Some of this concern may be alleviated by data showing that neither the number of needle passes nor the dwell time of a coaxial needle are associated with an increased incidence of pneumothorax.18,19 We also observed a reluctance of operators to perform CNBs, usually due to
Am J Clin Pathol 2015;143:193-200 197 DOI: 10.1309/AJCPMY8UI7WSFSYY
❚Table 2❚ Adequacy of FNAs for Molecular Testing by Operator Performing the Biopsy Operator (No. of FNAs Performed)
% of FNAs With Sufficient Material for Molecular Testing
P Valuea
1 (1) 2 (1) 3 (47) 4 (3) 5 (1) 6 (9) 7 (2) 8 (11) 9 (1) 10 (2) 11 (42)
0 0 63 3 100 11 0 54 100 0 38
NS NS .017 NS NS NS NS NS NS NS NS
FNA, fine-needle aspiration; NS, not significant. a Wald test applied to binomial logistic regression.
the risk of creating a pneumothorax.20 The true risk of pneumothorax may be difficult to estimate for operators, and use of smaller core needles can reduce that risk.21,22 Our comparison shows that there are significant differences between operators and/or biopsy technique. With some FNA operators, patients had a greater than 50% chance of requiring a repeat biopsy if molecular testing was needed. Published data on operator variability in image-guided biopsies are essentially nonexistent, although they are likely subject to confidential departmental quality assurance. Our numbers have to be interpreted with caution since operator performance is influenced by numerous variables, including patient characteristics, location and depth of the targeted lesion, fidelity and imaging speed of the CT scanner, and diameter and tip design of the biopsy needles. Most important, our comparison shows that even the most skilled FNA operator did not reach the adequacy rate of CNBs. We acknowledge several limitations of this study. Our institution prefers to perform molecular testing on cytology specimens through a similar platform used for our surgical pathology samples. Therefore, we are only comparing cell block preparations from FNA material with paraffinembedded CNB specimens. Cell block preparations can be cell poor despite adequately cellular aspirates, and higher success rates could possibly be achieved by scraping tumor cells directly from cytology smears.4,7 However, we prefer to leave cytology slides intact, especially if tumor cells are scant. Pathology practices submitting their tissue samples to referral laboratories for testing may also be less inclined to give up their original diagnostic cytology smears. Introducing ALK immunohistochemistry as a surrogate marker for ALK rearrangement could possibly lower the rate of insufficient ALK FISH samples.23 We also did not control for patient age, biopsy risk factors, or location of the targeted mass. We suspect that such variations were evenly distributed and differences negligible in our study
198 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
Probability of a Sufficient FNA Sample
Schneider et al / Lung Cancer Molecular Testing
1.0 0.8 0.6 0.4 0.2 0.0 0.0
2.0
4.0 6.0 Tumor Size (cm)
8.0
10.0
❚Figure 2❚ Regression curve for the binomial logistic model shows fine-needle aspirations (FNAs) more likely to yield sufficient samples for molecular testing if tumors are larger. Logistic function: Probability of a sufficient FNA sample = 1 – {1/[1 + exp(–1.342 + (0.376 × tumor size))]}.
population that included consecutive biopsy specimens diagnosed as adenocarcinoma or non–small cell carcinoma, favor adenocarcinoma. However, we cannot exclude that patients who underwent CNBs were “easier to biopsy.” We suspect that intratumor variability of oncogenic mutations may not be very common as initial reports suggested. Recent studies confirmed homogeneity of driver mutations in primary and metastatic lung carcinomas. For example, Yatabe et al24 suggested the concept of pseudoheterogeneity of EGFR mutations. They found identical EGFR driver mutations in multiple areas of the same tumor. However, the presence of EGFR amplification and/or contamination by normal cells may interfere with the interpretation of PCR mutation assays.24 In our setting, the main issue would be the detection limit when mutated tumor cells are admixed with nonmutated tumor cells. Taniguchi et al25 showed that EGFR mutations can still be detected in areas that contain nonmutated cells. Vignot et al26 were able to identify the driver mutation in lung cancer even if a tumor had accumulated other molecular alterations. Different pathologists could potentially influence the success rate of FNAs by not requesting testing on cell blocks that contain sufficient material. This seemed unlikely because all pathologists were educated on the minimum number of cells required by the laboratory. Also, since pathologists are aware of the importance of molecular testing, they tend to err on the side of caution and are more likely to request testing on cell blocks or CNB specimens that later turn out to be insufficient. Even if our minimum number of cells were too high, it should not influence the direct comparison between CNB and FNA specimens in this study.
© American Society for Clinical Pathology
AJCP / Original Article
The present study is unique because during the study period, we performed reflex molecular testing on eligible patients with lung carcinoma. Therefore, we can use cases that were eligible for testing but did not contain sufficient material for calculation of our adequacy rate. Most studies reporting high success rates of molecular testing do not consider in their calculation of success rates those biopsy specimens on which testing was not even attempted. Furthermore, to our knowledge, this is the only study to date that compares directly the yield of CNB and FNA specimens for EGFR, KRAS, and ALK testing in the same clinical environment. Although the 67% success rate of CNBs still leaves room for improvement, not being able to offer patients molecular testing on more than half of the FNA specimens is undesirable. For clinical environments in which only paraffin-embedded tissue is used for molecular testing of lung cancer, our data suggest that (1) CNBs are preferable over FNAs to obtain adequate tissue for molecular testing of lung adenocarcinomas, and (2) FNAs are more likely to yield sufficient samples from nodules above a certain size. Address reprint requests to Dr Schneider: Dept of Pathology, University of Pittsburgh Medical Center, 200 Lothrop St, Pittsburgh, PA 15213; [email protected].
References 1. Lindeman NI, Cagle PT, Beasley MB, et al. Molecular testing guideline for selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: guideline from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology. J Thorac Oncol. 2013;8:823-859. 2. Rekhtman N, Brandt SM, Sigel CS, et al. Suitability of thoracic cytology for new therapeutic paradigms in non-small cell lung carcinoma: high accuracy of tumor subtyping and feasibility of EGFR and KRAS molecular testing. J Thorac Oncol. 2011;6:451-458. 3. Boldrini L, Gisfredi S, Ursino S, et al. Mutational analysis in cytological specimens of advanced lung adenocarcinoma: a sensitive method for molecular diagnosis. J Thorac Oncol. 2007;2:1086-1090. 4. Billah S, Stewart J, Staerkel G, et al. EGFR and KRAS mutations in lung carcinoma: molecular testing by using cytology specimens. Cancer Cytopathol. 2011;119:111-117. 5. Smouse JH, Cibas ES, Janne PA, et al. EGFR mutations are detected comparably in cytologic and surgical pathology specimens of nonsmall cell lung cancer. Cancer. 2009;117:67-72. 6. Otani H, Toyooka S, Soh J, et al. Detection of EGFR gene mutations using the wash fluid of CT-guided biopsy needle in NSCLC patients. J Thorac Oncol. 2008;3:472-476. 7. Knoepp SM, Roh MH. Ancillary techniques on direct-smear aspirate slides: a significant evolution for cytopathology techniques. Cancer Cytopathol. 2013;121:120-128. 8. da Cunha Santos G, Liu N, Tsao MS, et al. Detection of EGFR and KRAS mutations in fine-needle aspirates stored on Whatman FTA cards: is this the tool for biobanking cytological samples in the molecular era? Cancer Cytopathol. 2010;118:450-456.
© American Society for Clinical Pathology
9. Travis W, Brambilla E, Noguchi M, et al. International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society international multidisciplinary classification of lung adenocarcinoma. J Thorac Oncol. 2011;6:244-285. 10. Bozzetti C, Negri FV, Azzoni C, et al. Epidermal growth factor receptor and Kras gene expression: reliability of mutational analysis on cytological samples. Diagnostic Cytopathol. 2013;41:595-598. 11. Ferretti GR, Busser B, de Fraipont F, et al. Adequacy of CT-guided biopsies with histomolecular subtyping of pulmonary adenocarcinomas: influence of ATS/ERS/IASLC guidelines. Lung Cancer. 2013;82:69-75. 12. Arcila ME, Oxnard GR, Nafa K, et al. Rebiopsy of lung cancer patients with acquired resistance to EGFR inhibitors and enhanced detection of the T790M mutation using a locked nucleic acid–based assay. Clin Cancer Res. 2011;17:1169-1180. 13. Solomon SB, Zakowski MF, Pao W, et al. Core needle lung biopsy specimens: adequacy for EGFR and KRAS mutational analysis. AJR Am J Roentgenol. 2010;194:266-269. 14. Cai G, Wong R, Chhieng D, et al. Identification of EGFR mutation, KRAS mutation, and ALK gene rearrangement in cytological specimens of primary and metastatic lung adenocarcinoma. Cancer Cytopathol. 2013;121:500-507. 15. Hsiao SH, Chung CL, Lee CM, et al. Suitability of computed tomography–guided biopsy specimens for subtyping and genotyping of non–small-cell lung cancer. Clin Lung Cancer. 2013;14:719-725. 16. Khan KA, Zaidi S, Swan N, et al. The use of computerised tomography guided percutaneous fine needle aspiration in the evaluation of solitary pulmonary nodules. Ir Med J. 2012;105:50-52. 17. Santambrogio L, Nosotti M, Bellaviti N, et al. CT-guided fine-needle aspiration cytology of solitary pulmonary nodules: a prospective, randomized study of immediate cytologic evaluation. Chest. 1997;112:423-425. 18. Halloush RA, Khasawneh FA, Saleh HA, et al. Fine needle aspiration cytology of lung lesions: a clinicopathological and cytopathological review of 150 cases with emphasis on the relation between the number of passes and the incidence of pneumothorax. Cytopathology. 2007;18:44-51. 19. Ko JP, Shepard JO, Drucker EA, et al. Factors influencing pneumothorax rate at lung biopsy: are dwell time and angle of pleural puncture contributing factors? Radiology. 2001;218:491-496. 20. Yeow KM, Su IH, Pan KT, et al. Risk factors of pneumothorax and bleeding: multivariate analysis of 660 CT-guided coaxial cutting needle lung biopsies. Chest. 2004;126:748-754. 21. Chakrabarti B, Earis JE, Pandey R, et al. Risk assessment of pneumothorax and pulmonary haemorrhage complicating percutaneous co-axial cutting needle lung biopsy. Respir Med. 2009;103:449-455. 22. Geraghty PR, Kee ST, McFarlane G, et al. CT-guided transthoracic needle aspiration biopsy of pulmonary nodules: needle size and pneumothorax rate. Radiology. 2003;229:475-481. 23. Yi ES, Chung JH, Kulig K, et al. Detection of anaplastic lymphoma kinase (ALK) gene rearrangement in non–small cell lung cancer and related issues in ALK inhibitor therapy: a literature review. Mol Diagn Ther. 2012;16:143-150. 24. Yatabe Y, Matsuo K, Mitsudomi T. Heterogeneous distribution of EGFR mutations is extremely rare in lung adenocarcinoma. J Clin Oncol. 2011;29:2972-2977.
Am J Clin Pathol 2015;143:193-200 199 DOI: 10.1309/AJCPMY8UI7WSFSYY
Schneider et al / Lung Cancer Molecular Testing
25. Taniguchi K, Okami J, Kodama K, et al. Intratumor heterogeneity of epidermal growth factor receptor mutations in lung cancer and its correlation to the response to gefitinib. Cancer Sci. 2008;99:929-935.
200 Am J Clin Pathol 2015;143:193-200 DOI: 10.1309/AJCPMY8UI7WSFSYY
26. Vignot S, Frampton GM, Soria JC, et al. Next-generation sequencing reveals high concordance of recurrent somatic alterations between primary tumor and metastases from patients with non–small-cell lung cancer. J Clin Oncol. 2013;31:2167-2172.
© American Society for Clinical Pathology
