ARMS for EGFR mutation analysis of cytologic and corresponding lung adenocarcinoma histologic specimens

Abstract

Purpose

Epidermal growth factor receptor (EGFR) mutation analysis is becoming a routine clinical practice for lung adenocarcinoma patients. Most patients with lung cancer are diagnosed at an advanced stage of the disease and are not suitable for surgical therapy. In many cases, cytologic specimens may be the only tissue available for diagnostic and molecular testing. Therefore, it is important to determine what condition the cytologic specimens should be in for adequately analyzing EGFR mutation status.

Methods

Fifty-eight paired cytologic and lung adenocarcinoma histologic specimens that satisfied 3 requisite parameters (>2 ng/μl DNA concentration, >30 tumor cells content, and >25 % tumor percentage) were collected. Exons 18 through 21 of the EGFR gene were analyzed by amplification refractory mutations system.

Results

The EGFR mutation concordance rate between cytologic specimens and corresponding histologic specimens was 100 %. A set of 30 paired specimens from different sites (lung and pleural fluids) from the same patient exhibited 100 % concordance in the EGFR mutation analysis results.

Conclusions

In this study, EGFR mutation presented a convenient and reliable method for the analysis of cytologic specimens that satisfied 3 requisite parameters (>2 ng/μl DNA concentration, >30 tumor cells content, and >25 % tumor percentage). We concluded that the specimens from both primary lung adenocarcinomas and metastatic lesions (such as pleural fluids) can be used for EGFR mutation analysis.

Introduction

Lung cancer has the highest incidence and is the most frequent cause of mortality worldwide among major cancers (Boyle and Levin 2008). Adenocarcinoma is the most common histologic subtype of lung cancer, accounting for almost half of all lung cancer cases (Curado et al. 2007). Because the epidermal growth factor receptor (EGFR) mutation is a validated predictive marker for response and overall survival time with EGFR tyrosine kinase inhibitors in advanced lung adenocarcinoma, experts have recommend that patients with advanced adenocarcinomas be tested for the EGFR mutation. Approximately 70 % of advanced-stage lung cancers are diagnosed and staged using small biopsies or cytology (including fine-needle aspiration, pleural fluid, and bronchial washing or brushing specimens) rather than surgical resection specimens (Shah et al. 2006), which may be the only material available for EGFR mutation analysis. A recent review demonstrated that the most common types of cytologic specimens for assessing EGFR mutations were formalin-fixed, paraffin-embedded cell blocks, archival smear slides and fresh cells (da Cunha Santos et al. 2011). Cytologic material can often only be obtained in small amounts via fine-needle aspiration and is not suitable to be made into cell blocks; however, smear slides are easily made, so cytologic material is always examined using smear slides. Therefore, it is important to determine what conditions are adequate for analysis of cytologic specimens to evaluate EGFR mutation status.

Several studies have demonstrated the suitability of cytological specimens for detecting the EGFR mutation (Tanaka et al. 2007; Malapelle et al. 2012; Bruno et al. 2011; Aisner et al. 2013). However, few of those studies have confirmed the optimal conditions for cytologic specimens, have compared cytologic specimens with histologic specimens from the same patient, and have compared the mutation status of primary tumors with metastatic lesions using cytologic specimens. In the current study, we compared 58 cytologic smears with formalin-fixed, paraffin-embedded (FFPE) specimens, including biopsies and resections from lung adenocarcinomas, and determined which of these cytologic specimens were suitable for the molecular assessment of EGFR mutation status. The condition of each of the patients was also confirmed.

Materials and methods

Sample collection and processing

This study was approved by the Institutional Review Board of our hospital. We retrospectively reviewed and assessed the EGFR mutation status of 58 surgically resected or biopsied primary or metastatic lung adenocarcinomas, confirmed by morphology and immunohistochemistry with available cytologic specimens collected at our hospital between October 2013 and March 2014. These specimens included 18 surgically resected lung tissues, 10 surgically resected regional lymph nodes and 30 biopsied lung tissues. The 58 cytologic archival smear slides from these 58 patients were examined, including 18 fine-needle aspirations of the lung, 10 fine-needle aspirations of regional lymph nodes and 30 pleural fluid samples. Cytomorphological evaluation of smears was conducted in all cases, confirming the diagnosis.

Specimen selection for EGFR mutation testing

All cytologic archival smear slides were independently reviewed by 2 pathologists. The slides were assessed for the number of tumor cells, and there were at least 30 tumor cells within each slide for selection. The tumor percentage was also calculated based on the number of tumor cells relative to all nucleated cells within each slide, and the lowest tumor percentage allowed for detection was 25 %. The histologic specimens were evaluated using the same method.

DNA extraction and ARMS mutation analysis

Genomic DNA was manually isolated using a QIAamp DNA FFPE Tissue Kit from FFPE sections and smear slides (Qiagen, Germany) according to the manufacturer’s instructions. DNA was isolated by elution with 50 μl of Tris/Acetate/EDTA(TAE), and the genomic DNA was then quantified using Nano UV spectrophotometer. OD 260/280 values should be between 1.8 and 2.0. An ADx-ARMS EGFR mutation detection kit was used to analyze the DNA template, which was adjusted to a concentration of 2 ng/μl on a real-time PCR instrument (Stratagene Mx3005P) according to the manufacturer’s instructions. The results were interpreted by an experienced technician. Both DNA extraction and ARMS method for EGFR analysis were previously validated in our laboratory. Positive and negative controls were analyzed routinely for lab procedure.

Results

The EGFR mutation test results from 58 cytologic and corresponding histologic specimens are provided in Table 1. The specimens included samples from 30 men and 28 women. Some cytologic archival smear samples had sufficient tumor cells so that only one slide was necessary. In those cases, the slide was divided into two parts: one part of the slide was scraped off using a scalpel for EGFR analysis, and the remainder was stored in the archive. The cytologic and histologic specimens were evaluated for the quality and quantity of tumor tissue available for EGFR testing and were assessed in terms of 3 factors: DNA concentration, tumor cell number and percentage of tumor cells. For ARMS, the lowest DNA concentration that still allowed for detection was 2 ng/μl. The DNA concentration extracted from the cytologic archival smear slides ranged from 2.1 to 578.8 ng/μl. The DNA concentration extracted from the FFPE specimens ranged from 3.4 to 626.8 ng/μl.

Table 1.

EGFR mutation status in cytologic and corresponding histologic specimens

No.	Sex	Age	Procedure site		HS type	CS type	DNA concentration (ng/µl)		EGFR mutation status		Treatment
			HS	CS			HS	CS	HS	CS
1	M	76	Lung	Pleura	FNB	Fluid	13.5	165.6	21L858R	21L858R	EGFR-TKI
2	M	44	Lung	Pleura	FNB	Fluid	28.6	261.4	–	–	Chemoradiotherapy
3	M	61	Lung	Pleura	FNB	Fluid	20.3	431.2	–	–	Chemotherapy
4	F	45	Lung	Pleura	FNB	Fluid	27.9	45.2	–	–	Chemoradiotherapy
5	F	48	Lung	Pleura	FNB	Fluid	38.1	225.1	21L858Ra	21L858R	EGFR-TKI
6	F	67	Lung	Pleura	FNB	Fluid	17.1	424.2	–	–	Chemoradiotherapy
7	F	37	Lung	Pleura	FNB	Fluid	9.5	33.1	21L858R	21L858R	EGFR-TKI
8	F	67	Lung	Pleura	FNB	Fluid	38.4	347.2	19del	19del	EGFR-TKI
9	F	59	Lung	Pleura	FNB	Fluid	4.6	108.0	21L858R	21L858R	EGFR-TKI
10	M	69	Lung	Pleura	FNB	Fluid	12.6	125.5	–	–	Chemotherapy
11	M	36	Lung	Pleura	FNB	Fluid	9.7	218.3	21L858R	21L858R	EGFR-TKI
12	M	56	Lung	Pleura	FNB	Fluid	6.4	48.0	19dela	19del	EGFR-TKI
13	M	66	Lung	Pleura	FNB	Fluid	12.7	46.0	–	–	Chemotherapy
14	M	61	Lung	Pleura	FNB	Fluid	6.2	28.7	–	–	Chemotherapy
15	M	77	Lung	Pleura	FNB	Fluid	23.5	311.0	–	–	Chemotherapy
16	F	66	Lung	Pleura	FNB	Fluid	16.7	150.3	19del	19del	EGFR-TKI
17	F	78	Lung	Pleura	FNB	Fluid	31.0	30.4	–	–	Chemotherapy
18	F	81	Lung	Pleura	FNB	Fluid	25.8	133.1	–	–	Chemotherapy
19	F	64	Lung	Pleura	FNB	Fluid	7.2	218.0	21L858R	21L858R	EGFR-TKI
20	F	79	Lung	Pleura	FNB	Fluid	3.4	16.1	21L858R	21L858R	EGFR-TKI
21	F	78	Lung	Pleura	FNB	Fluid	5.8	47.1	–	–	Chemotherapy
22	M	82	Lung	Pleura	FNB	Fluid	19.6	172.0	21L858R	21L858R	EGFR-TKI
23	M	54	Lung	Pleura	FNB	Fluid	18.2	229.9	–a	–	Chemoradiotherapy
24	F	54	Lung	Pleura	FNB	Fluid	8.3	344.7	–	–	Chemoradiotherapy
25	F	48	Lung	Pleura	FNB	Fluid	12.9	67.1	–	–	Chemoradiotherapy
26	F	56	Lung	Pleura	FNB	Fluid	7.6	40.7	–	–	Chemoradiotherapy
27	F	68	Lung	Pleura	FNB	Fluid	6.8	250.8	21L858R	21L858R	EGFR-TKI
28	M	62	Lung	Pleura	FNB	Fluid	9.5	215.2	–	–	Chemoradiotherapy
29	M	80	Lung	Pleura	FNB	Fluid	14.2	51.8	–	–	Chemotherapy
30	M	76	Lung	Pleura	FNB	Fluid	11.8	578.8	19del	19del	EGFR-TKI
31	F	74	Lung	Lung	Resection	FNA	534.1	22.1	19del	19del	Surgery + EGFR-TKI
32	F	71	Lung	Lung	Resection	FNA	309.5	2.1	19del, 20S768Ia	19del, 20S768I	Surgery + chemotherapy
33	M	60	Lung	Lung	Resection	FNA	442.9	4.0	–	–	Surgery + chemoradiotherapy
34	M	84	Lung	Lung	Resection	FNA	389.7	2.4	21L858R	21L858R	Surgery + EGFR-TKI
35	M	56	Lung	Lung	Resection	FNA	626.7	16.0	21L858R	21L858R	Surgery + EGFR-TKI
36	F	34	Lung	Lung	Resection	FNA	512.0	35.6	19del	19del	Surgery + EGFR-TKI
37	M	57	Lung	Lung	Resection	FNA	213.8	17.6	19del	19del	Surgery + EGFR-TKI
38	M	66	Lung	Lung	Resection	FNA	387.9	26.3	–	–	Surgery + Chemotherapy
39	M	70	Lung	Lung	Resection	FNA	358.4	33.2	19dela	19del	Surgery + EGFR-TKI
40	F	64	Lung	Lung	Resection	FNA	329.6	73.3	–a	–	Surgery + chemotherapy
41	M	72	Lung	Lung	Resection	FNA	471.8	78.3	–	–	Surgery
42	M	60	Lung	Lung	Resection	FNA	508.2	11.7	21L858Ra	21L858R	Surgery + EGFR-TKI
43	F	69	Lung	Lung	Resection	FNA	626.8	95.5	19del	19del	Surgery + EGFR-TKI
44	F	62	Lung	Lung	Resection	FNA	347.8	73.8	21L858R	21L858R	Surgery + EGFR-TKI
45	M	77	Lung	Lung	Resection	FNA	237.1	28.6	–	–	Surgery
46	F	67	Lung	Lung	Resection	FNA	416.5	8.7	21L858R	21L858R	Surgery + EGFR-TKI
47	F	64	Lung	Lung	Resection	FNA	374.1	17.3	21L858R	21L858R	Surgery + EGFR-TKI
48	F	41	Lung	Lung	Resection	FNA	298.5	79.9	–	–	Surgery + chemotherapy
49	F	77	LN	LN	Resection	FNA	115.2	171.5	–	–	Chemotherapy
50	M	62	LN	LN	Resection	FNA	226.7	109.8	19del	19del	EGFR-TKI + radiotherapy
51	M	62	LN	LN	Resection	FNA	231.8	95.6	19del	19del	EGFR-TKI + radiotherapy
52	M	79	LN	LN	Resection	FNA	221.5	31.1	–	–	Chemotherapy
53	M	65	LN	LN	Resection	FNA	310.7	269.3	21L858Ra	21L858R	EGFR-TKI + radiotherapy
54	M	57	LN	LN	Resection	FNA	162.1	256.1	–	–	Chemoradiotherapy
55	M	55	LN	LN	Resection	FNA	180.1	22.5	–	–	Chemoradiotherapy
56	F	37	LN	LN	Resection	FNA	175.3	38.6	19dela	19del	EGFR-TKI + radiotherapy
57	F	64	LN	LN	Resection	FNA	180.9	26.3	–	–	Chemoradiotherapy
58	M	54	LN	LN	Resection	FNA	215.7	89.0	–	–	Chemoradiotherapy

CS cytologic specimens, EGFR epidermal growth factor receptor, F female, FNA fine-needle aspiration, FNB fine-needle biopsy, HS histologic specimens, LN lymph node, M male, TKI tyrosine kinase inhibitor, WT wild type, 19del in-frame deletions in exon 19, 20S768I S768I missense mutation in exon 20, 21L858R a point mutation L858R in exon 21

^aThe results were validated by direct sequencing

The EGFR mutation was found in 13 of the 30 cytologic archival smear slides of pleural fluid (Fig. 1), including 4 exon 19 deletions and 9 exon 21 L858R substitutions. 12 of the 18 fine-needle aspiration smears of the lung were positive for the EGFR mutation, including 6 deletions in exon 19 (one case with exon 20 S768I) and 6 exon 21 L858R substitutions. 3 of the 10 fine-needle aspiration regional lymph node smears showed exon 19 deletions, and 1 of 10 had exon 21 L858R substitutions. In total, 29 of the 58 cytologic archival smear slides were positive for EGFR mutation (Fig. 2). Exon 21 L858R substitutions were the most common (16 of 29 patients), while exon 19 deletions (13 of 29 patients) and exon 20 mutation (1 of 29 patients) were the next most frequent mutations among the remaining patients. There was 1 patient who had a mutation in both exon 19 and exon 20.

Fig. 1.

A cytologic archival smear of pleural fluid shown, low power (a) and high power (b)

Fig. 2.

Detection of epidermal growth factor receptor (EGFR) mutation shown by amplification refractory mutations system (ARMS). a Exon 19 deletion mutation as identified by ARMS. b An exon 19 deletion and an exon 20 point mutation as detected by ARMS. c ARMS identification of an exon 21 point mutation

These specimens all fulfilled the designated criteria (>2 ng/μl DNA concentration, >30 tumor cells content, and >25 % tumor percentage), and the cytologic specimens exhibited 100 % concordance with the corresponding histologic specimens. The concordance rate of EGFR mutations between the cytologic specimens and the corresponding histologic specimens was 100 % based on EGFR mutation assessment performed on cytologic specimens of metastatic lesions compared with histologic specimens of primary tumor sites from same patients.

Discussion

Epidermal growth factor receptor is a member of the ErbB (erythroblastic leukemia viral oncogene homolog) family of transmembrane tyrosine kinase receptor proteins. Activating EGFR mutations could lead to uncontrolled cell proliferation and resistance to chemotherapy (D’Angelo et al. 2011a, b; Gupta et al. 2009). The incidence of EGFR mutations varies by gender, ethnicity and smoking status (D’Angelo et al. 2011a, b; Girard et al. 2012; Reinersman et al. 2011). The two most common EGFR mutations in lung adenocarcinomas, which account for 90 % of the mutations, are in-frame deletions in exon 19 and a point mutation at codon 858 in exon 21 (L858R) (Ladanyi and Pao 2008).

EGFR mutations were tested using various technologies, including polymerase chain reaction (PCR) amplification and sequencing, an amplification refractory mutations system (ARMS), and peptide nucleic acid-locked PCR, among others (Pao and Ladanyi 2007; Ellison et al. 2010). According to several previous studies, ARMS is a more sensitive and robust assay for the detection of somatic mutations than standard DNA sequencing and can detect as few as 1 % of tumor cells in lung cancer (Ellison et al. 2010; Kimura et al. 2006). In the present study, we evaluated the presence of EGFR mutations in cytologic smear slides and corresponding histologic specimens using ARMS. Scalpel extraction is a relatively simple macrodissection method (Khode et al. 2013). Chowdhuri et al. (2012) demonstrated the feasibility of using laser-capture microdissection for EGFR testing of cytologic samples; however, laser-capture microdissection is an expensive method that is not available to all laboratories.

DNA concentration, tumor cell number, and percentage of tumor cells were important and could interfere with obtaining sufficient material for EGFR analysis from the cytologic samples. This difficulty has been highlighted in several studies, including one conducted by Savic et al. (2008), who determined that 30 cells was the minimum number required for successful DNA sequencing analysis of EGFR. Furthermore, Smouse et al. (2009) reported the detection of mutations in cytologic samples with tumor percentages between 25 and 50 % using a direct sequencing method. A matched group of cytologic and corresponding histologic specimens was analyzed by Sun et al. (2013), who demonstrated that the minimal requirements for cytologic samples that allowed for successful pyrosequencing analysis of EGFR were: a DNA concentration >25 ng/μl, the presence of >30 tumor cells, or a tumor percentage >30 %. They found that cytologic specimens that satisfied at least 1 of these 3 requirements exhibited 100 % concordance with the corresponding histologic specimens. On the basis of these results, our study determined the EGFR mutation status of fifty-eight paired cytologic and histologic specimens of lung adenocarcinomas that satisfied the 3 parameters (>2 ng/μl DNA concentration, >30 tumor cells content, and >25 % tumor percentage) as analyzed by ARMS. The cytologic specimens exhibited 100 % concordance with the corresponding histologic specimens. To validate the results obtained by ARMS, the EGFR mutations status of 9 cases of histologic specimens (as shown in Table 1) was detected by direct sequencing, which had shown 100 % concordance with ARMS. We recommend using the ARMS method to evaluate EGFR mutation in cytologic samples that satisfy the 3 requisite parameters.

In the present study, the majority of cytologic specimens from metastatic lesions were pleural fluids (30 of 40 specimens), and the majority of histologic specimens from primary tumors were obtained by FNB (30 of 48 specimens). The concordance rate of EGFR mutations from different sites (cytologic specimens from metastatic lesions and histologic specimens from primary tumors) from the same patient was 100 %. Several previous studies have reported a heterogeneous distribution of EGFR mutations in primary lung adenocarcinomas and corresponding metastatic lesions (Sun et al. 2011; Kalikaki et al. 2008; Han et al. 2011; Navani et al. 2012). This difference may be explained by the following 2 factors: (1) the cases of matched cytologic and histologic specimens were limited, and (2) heterogeneity in the distribution of EGFR mutations is rare in lung adenocarcinomas, as noted by Yatabe et al. (2011).

In conclusion, EGFR mutation analysis of cytologic specimens that satisfy 3 designated parameters (>2 ng/μl DNA concentration, >30 tumor cells content, and >25 % tumor percentage) is a convenient and reliable procedure. We suggest that specimens from both primary lung adenocarcinomas and metastatic lesions could be used for EGFR mutation analysis.