Article Text

Download PDFPDF

EGFR mutation detection by microfluidic technology: a validation study
  1. Umberto Malapelle1,
  2. Stefania Russo1,
  3. Francesco Pepe1,
  4. Roberta Sgariglia1,
  5. Caterina De Luca1,
  6. Claudio Bellevicine1,
  7. Pierlorenzo Pallante2,
  8. Giancarlo Troncone1
  1. 1Dipartimento di Sanità Pubblica, Università di Napoli Federico II, Napoli, Italia
  2. 2Consiglio Nazionale delle Ricerche, Istituto di Endocrinologia ed Oncologia Sperimentale, Napoli, Italia
  1. Correspondence to Professor Giancarlo Troncone, Dipartimento di Sanità Pubblica, Università di Napoli Federico II, via Sergio Pansini 5, Napoli I-80131, Italy; giancarlo.troncone{at}


Advanced non-small cell lung cancer samples are tested for epidermal growth factor receptor (EGFR) gene mutations. Their detection by direct sequencing is time-consuming. Conversely, the length analysis of fluorescently labelled PCR products is easier. To avoid labelled primers and the automated capillary electrophoresis apparatus, we validated a fast and sensitive chip-based microfluidic technology. The limit of detection of fragment length assay on microfluidic device was 5%, more sensitive than direct sequencing (12.5%). The novel methodology showed high accuracy in the analysis of samples whose mutational status was known. The accuracy in quantifying mutated alleles (mA) was evaluated by PCR products subcloning; the mA% provided by direct sequencing of subcloned PCR products showed a close correlation with the mA% provided by the microfluidic technology for both exon 19 (R2=0.9) and 21 (R2=0.9). Microfluidic-based on-chip electrophoresis makes EGFR testing more rapid, sensitive and cost-effective.

  • EGFR
  • LUNG

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


Epidermal growth factor receptor (EGFR) gene mutations predict sensitivity to the administration of EGFR inhibitors. Short in-frame deletions in exon 19 and the L858R exon 21 mutation comprise up to 90% of all activating EGFR mutations.1 Traditionally, their detection is based on the most widely available technique, that is, PCR followed by direct sequencing.2 This latter offers the advantage of detecting all mutations, but it is time-consuming and has low sensitivity.2 Conversely, fragment length analysis detects the exon 19 alterations easily3; the deleted allele generates in the electropherogram an additional peak, which is detected by the use of labelled primers on an automated capillary electrophoresis (CE) apparatus (GeneScan; Applied Biosystems, Foster City, California, USA). Similarly, the exon 21 L858R is detected by CE analysis of fluorescently labelled PCR products digested by restriction fragment length polymorphism (RFLP) analysis.3 The rapidity and ease of this approach led several institutions to adopt it in their practice.3–5 Here we show that microfluidic technology can make fragment length analysis of EGFR even more rapid, sensitive and cost-effective. Dye molecules intercalated into DNA are detected by laser-induced fluorescence with high-resolution capacity, avoiding the need of fluorescently labelled primers.

Materials and methods

Cell lines

PC9 cells that harbour a Glu746–Ala750 deletion in exon 19, H1975 cells that carry the L858R point mutation, and the wild-type (WT) A549 cells were employed to assess the lowest limit of detection (LOD) of EGFR mutant alleles by fragment length assay performed with microfluidic technology. This was compared with the LOD by fragment length assay performed by traditional CE and with standard direct sequencing. The A549 and H1975 cell lines were obtained from American Type of Culture Collection (Rockville, Maryland, USA). The PC9 cell line was obtained from the CNR/IEOS Institute (Naples, Italy). To determine the LOD, we serially mixed the mutated (PC9 and H1975) and WT (A549) EGFR cell lines at dilutions of 50%, 25%, 12.5%, 10% and 5%.

Clinical samples

We retrieved from our archive a sizeable number (n=20) of non-small cell lung cancer (NSCLC) DNA samples, derived from either histological biopsies (n=10) or cytological (n=10) specimens, which had previously been tested for EGFR mutation by fragment length assay (exon 19) and by RFLP (exon 21). Twelve samples carried EGFR mutations (exon 19 del, n=6; L858R n=6). Written consent was received by all patients.

Fragment length analysis by microfluidic technology

Primers and PCR conditions for exons 19 and 21 are detailed elsewhere.3 ,6 1 µl of each PCR products were prepared for electrophoresis on a integrated micro-channels chip (BIORAD DNA 1-K microchip). Eleven samples were simultaneously separated by a single chip according to the manufacturer's instructions on Experion electrophoresis instrument (BIORAD, Milan, Italy).7 The data were analysed by the Experion software. The results were displayed as both simulated gel images and electropherograms. All analyses were performed in duplicate and confirmed by direct sequencing as previously described.6 Peaks relative to the WT and mutant alleles were quantified by the Experion software, and concentration data were reported as nanogram per microlitre.7 To calculate the mutant allele percentage, we used the following formula: (mut)/((mut)+(wt)) (%). The accuracy of this approach was assessed by comparing the mutant allele percentage with that obtained by subcloning and sequencing the PCR products; to this end, for each sample, 30 plasmids were cloned into a topoisomerase (TOPO) TA vector (Invitrogen, California, USA) according to the manufacturer's instructions.8 Finally, we evaluated the relationship between the mutated alleles (mA)% provided by microfluidic technology and that of direct sequencing of subcloned PCR products by a linear regression model.


Microfluidic technology on cell lines and clinical samples

The LOD of fragment length assay by microfluidic technology was 5% for mutant exons 19 and 21 alleles (figure 1A); LOD was higher for fragment length assay by standard CE (10%; figure 1B) and for direct sequencing (12.5%; figure 1C). In all histological and cytological samples, whose mutational status was known (n=12), the presence and type of mutation were confirmed.

Figure 1

The limit of detection (LOD) of mutant epidermal growth factor receptor alleles assessed by fragment length assay by microfluidic technology was 5% for exons 19 and 21 (A). Fragment length assay by standard capillary electrophoresis showed a 10% LOD (B). Direct sequencing LOD was 12.5% (C).

Exon 19 mutant allele dosage quantified by microfluidic technology ranged between 0.2 ng/µl and 2.4 ng/µl, and exon 21 dosage ranged between 0.1 ng/µl and 1.2 ng/µl. For each single case, the mA% derived by microfluidic technology and subcloning sequencing is reported in table 1. Exon 19 mA% correlation (R2) was 0.9275, and that relative to Exon 21 mA% was 0.9100 (figure 2).

Table 1

Data relative to mA% assessed by microfluidic technology and subcloning sequencing for exons 19 (cases 1–6) and 21 (cases 7–12)

Figure 2

Graphic representation of the linear correlation between the mA% derived by microfluidic technology (y-axis) and subcloning sequencing (x-axis) for exons 19 (left, R2=0.9275) and 21 (right, R2=0.9100).


Here we demonstrated the feasibility and the accuracy of microfluid technology to analyse the mutational status of EGFR in histological and cytological samples of NSCLC. In addition, this methodology reliably quantifies the mutant allele dosage. To date, length analysis of EGFR PCR has required the use of labelled primers and of an automated CE, both increasing the expense of the technique. Our results suggest that fragment length analysis to detect exon 19 deletions and exon 21 L858R is less time-consuming and simpler if performed by microfluidic technology.

In Europe, EGFR testing is mandatory for the prescription of gefitinib in NSCLC patients. However, not all pathology departments are equipped to run molecular diagnostics; there is the need to develop simple, fast and cheap molecular assays. Here we show that a novel and relatively low-cost microfluidic apparatus (BioRad Experion) strongly reduces the handling procedures, the separation time and sample consumption compared with standard CE. The chip preparation and loading take only a few minutes and there is room for 11 samples. In addition, quantitative data (DNA fragment size and concentration) are obtained after each run. Run information data are reported as virtual gel bands (figure 1) and peaks, whose analysis does not require specific skills.

Approximately 20% of lung adenocarcinomas in western populations are characterised by EGFR gene copy number gain (CNG).9 Since the ratio between mutant and WT alleles is maintained after transcription, the mutant allele-specific imbalance (MASI) may be a critical biological predictor of tumour behaviour.9 However, CNG may be difficult to quantify when DNA from contaminating normal tissue is also analysed.9 Data relative to mA% by microfluidic technology closely correlated with those obtained by exons 19 (R2=0.9275) and 21 (R2=0.9100) subcloning sequencing. Thus, our assay is a simple and accurate tool to evaluate EGFR MASI in clinical samples.

Key messages

  • Fragment length assay, one of the methods most commonly used in clinical practice to assess EGFR mutational status, is usually performed by conventional capillary electrophoresis.

  • Microfluidic technology can make fragment length assay more rapid, sensitive and cost-effective.

  • Microfluidic technology is a very accurate tool to quantify the amount of mutated allele.



  • Contributors UM conceived the study. SR, FP, RS and CdL performed the experiments. CB contributed as referral pathologist. PP performed subcloning experiments. GT supervised the study and wrote the manuscript with the contribution of the other authors. GT is the guarantor of the paper. All authors read and approved the final version.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval 185/10.

  • Provenance and peer review Not commissioned; externally peer reviewed.