pubs.acs.org/ac Article One-Step and Real-Time Detection of microRNA-21 in Human Samples for Lung Cancer Biosensing Diagnosis Olalla Calvo-Lozano, Pablo García-Aparicio, Lajos-Zsolt Raduly, Maria Carmen Estévez,* Ioana Berindan-Neagoe, Manuela Ferracin, and Laura M. Lechuga* Cite This: Anal. Chem. 2022, 94, 14659−14665 Read Online ACCESS Metrics & More Article Recommendations * sı Supporting Information ABSTRACT: The rapid diagnosis of cancer, especially in its early stages, is crucial for on-time medical treatment and for increasing the patient survival rate. Lung cancer shows the highest mortality rate and the lowest 5-year survival rate due to the late diagnosis in advanced cancer stages. Providing rapid and reliable diagnostic tools is a top priority to address the problem of a delayed cancer diagnosis. We introduce a nanophotonic biosensor for the direct and real-time detection in human plasma of the microRNA-21-5p biomarker related to lung cancer. The biosensor employs a silicon photonic bimodal interferometric waveguide that provides a highly sensitive detection in a label-free format. We demonstrate a very competitive detectability for direct microRNA-21-5p biomarker assays in human plasma samples (estimated LOD: 25 pM). The diagnostic capability of our biosensor was validated by analyzing 40 clinical samples from healthy individuals and lung cancer patients, previously analyzed by reverse-transcription quantitative polymerase chain reaction (qRT-PCR). We could successfully identify and quantify the levels of microRNA in a one- step assay, without the need for DNA extraction or amplification steps. The study confirmed the significance of implementing this biosensor technique compared to the benchmarking molecular analysis and showed excellent agreement with previous results employing the traditional qRT-PCR. This work opens new possibilities for the true implementation of point-of-care biosensors that enable fast, simple, and efficient early diagnosis of cancer diseases. L ung cancer (LC) presents the highest incidence and mortality rates worldwide. In 2020, 2.206.771 new cases and 1.796.144 deaths were reported, representing close to 1 in even in the early stages, before tumor presence, by employing a noninvasive and inexpensive method5 permitting rapid medical treatment and increasing the cancer survival rate. To diagnose 5 (18.4%) cancer-related deaths.1 In the last few years, LC has LC, a large list of different biomarkers has been described such caused more deaths than breast, prostate, colorectal, and brain as proteins (i.e., NSE, CEA, and CYFR A-21),6 epigenetic cancers combined, being the top cancer death in men and the events like DNA methylation (CDO1 gene, ZNF177 gene),7 second one in females, after breast cancer. Besides, LC has a 5- DNA mutations (K-RAS gene, PTEN gene),8 or circulating year survival rate of 19%, second only to pancreatic cancer, microRNAs (microRNA-21-5p, microRNA-205-5p, and micro- which is the cancer with the poorest prognosis.2 Most LCs are RNA-210-3p) among others.9 Specifically, microRNAs, which asymptomatic (or present common symptoms such as cough, are short and single-stranded noncoding RNAs (∼22 anorexia, fatigue, or dyspnea).3 Thus, individuals are not nucleotides), play an important role in the modulation of diagnosed promptly, leading to long-term sequels like several biological processes, such as cell cycle control, advanced cancer stages characterized by metastasis in which apoptosis, and differentiation, being involved in the tumori- treatment is not able to effectively injure the tumor.4 genesis process.9 The expression of a specific microRNA Nowadays, the techniques for LC diagnosis rely on imaging signature can determine the type of cancer according to the methodologies like chest radiography, contrast-enhanced tissue10 and its development and progression stage (early or computed tomography (CT), and positron emission tomog- late stage, metastasis...).11 At present, microRNA detection is raphy. These techniques show low sensitivity, are costly, and based on reverse-transcription quantitative polymerase chain imply radiation generation.5 In addition, they are only useful when the tumor is visible enough. For optimizing the diagnosis and treatment, they should be combined with other molecular Received: July 5, 2022 techniques such as cytology samples and small biopsies.3 Accepted: September 27, 2022 Clinical diagnosis based on detecting biomarkers such as Published: October 11, 2022 proteins or genetic material in body fluids such as urine, saliva, or blood is boosting the cancer diagnosis. The biomolecular analysis enables the identification of a cancerous development © 2022 The Authors. Published by American Chemical Society https://doi.org/10.1021/acs.analchem.2c02895 14659 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article Table 1. Nucleotide Sequences Employed in This Work nucleotide sequences sequence (5′ → 3′) DNA probe miRNA-21 [Thiol]TTT TTT TTT TTT TTT TCA ACA TCA GTC TGA microRNA-21-5p (target) UAGCUUAUCAGACUGAUGUUGA microRNA-210-3p (control) CUGUGCGUGUGACAGCGGCUGA reaction (qRT-PCR) or digital PCR techniques, Northern independent arms. 20 In addition, BiMW sensor chip Blot, and high-throughput sequencing (i.e., microarrays or next fabrication is performed by standard microelectronic technol- generation sequencing). Despite the reliability and feasibility of ogy that enables reducing fabrication costs and increasing these established techniques, they might require large amounts reproducibility and reliability. The BiMW biosensor has of purified sample, labels, and long incubation times that imply previously demonstrated its potential for clinical diagnostics a laborious preparation and rigorous experimental conditions. in several areas, including infectious diseases,21,22 cancer Moreover, the lack of standardized protocols for microRNA diagnosis,23 and endocrinology,24 enabling direct, sensitive, extraction from blood can introduce a high variability, leading and reliable detection of bacteria, microRNAs, and hormones, to noncomparable analyses.12 Therefore, efficient and reliable respectively. So far, we had not yet evaluated the effect of detection strategies for circulating microRNA quantification complex fluids like serum and plasma with this device. Herein, are crucial. Rapid and simple detection of microRNA we have implemented and validated this powerful biosensor for biomarkers in the blood of cancer patients can not only the detection of circulating microRNAs released in the blood facilitate prompt treatment but also potentially increase the of non-small cell lung cancer (NSCLC) patients. By survival rate and decrease the mortality rate. functionalizing the sensor surface with specific DNA probes, Biosensors constitute an excellent opportunity to develop we carried out a complementary hybridization assay to identify integrated devices that enable fast and accurate clinical the presence of a relevant microRNA biomarker (microRNA- diagnosis. In particular, optical evanescent wave-based 21-5p) present in plasma. The BiMW sensor enables not only technology offers label-free and real-time quantitative analysis direct and real-time quantitative analysis but also provides a with high sensitivity and a remarkable potential for rapid detection and diagnosis (less than 45 min) without the miniaturization in point-of-care devices.13 Evanescent wave need for amplification steps or sample pretreatment, contrary biosensors are excellent tools for the detection of microRNAs to the conventional qRT-PCR. This methodology represents a due to the minimal sample preparation, absence of significant step forward for the use of optical biosensor devices amplification steps, high sensitivity, fast outcome, and for complex fluid analysis and particularly for an efficient multiplexing capability. Several optical biosensors have been cancer diagnosis. proposed to detect circulating microRNAs for clinical diagnosis. The most employed biosensors are based on the surface plasmon resonance, reaching limits of detection (LOD) ■ EXPERIMENTAL SECTION Chemical and Biological Reagents. All chemical in the pM−fM range when they include amplification steps reagents are described in the Supporting Information (SI). with antibodies,14 gold nanoparticles,15 or catalytic reactions.16 The DNA probe and synthetic microRNAs employed for the Among silicon photonic biosensors, microring resonators17 optimization of the hybridization complementary assay were and a Mach Zehnder interferometer18 have been also purchased from IBIAN Technologies (Zaragoza, Spain) and employed for microRNA detection, reaching LOD in the nM are summarized in Table 1. Human pooled plasma was range. purchased from Innovative Research (Michigan, US). We present an advanced nanophotonic biosensor based on BiMW Biosensor Device. The biosensor employed is an bimodal waveguide interferometers (BiMW).19 The working in-house designed and assembled BiMW device that principle of a BiMW biosensor relies on the evanescent wave, incorporates all the optical and microfluidic components. A an electromagnetic field generated when polarized light detailed description is provided in the SI and Figure S1. propagates through a waveguide by total internal reflection. Biofunctionalization of the Sensor Chips with the The evanescent wave arises when part of the energy is not DNA Capture Probe. The sensor chips were cleaned and totally confined and penetrates in the external medium up to silanized with APTES-PDITC (as described in the SI) and hundreds of nanometers. The evanescent wave is highly placed on the experimental set-up for the in situ immobiliza- sensitive to changes in the refractive index of the medium and tion of the thiolated DNA probes to the R-NCS groups in close proximity to or on the waveguide’s surface. In the case (Figure S2A). DEPC-H2O water was kept as running buffer, of the BiMW biosensor, a polarized monochromatic light is and the immobilization solution was flown at a constant rate of coupled into the bimodal waveguides and propagated along its 3 μL min−1. The immobilization solution contained a mixture core, allowing the excitation of two light modes. These modes of a DNA SH-T15-miRNA21 probe and SH-PEG-COOH produce an interference pattern that is dependent on the local spacer (total thiol 2 μM, molar ratio of DNA/spacer of 1:1) in refractive index at the waveguide surface. Any event at the phosphate immobilization buffer. Before injection, the sensor surface, such as the binding of an analyte to its specific immobilization solution was incubated with 0.1 μM TCEP receptor, results in a change in the local refractive index, which solution in constant agitation for 20 min at 36 °C. produces a phase shift between the two modes, and hence, an To avoid nonspecific adsorptions from plasma samples, a interference pattern that can be monitored instantaneously. blocking step was included by employing BSA 20 mg mL−1 The BiMW design is based on a linear straight waveguide with diluted in PBS which was injected over the sensor chip after a step junction, its fabrication being simpler than the one the immobilization step at 5 μL min−1. Finally, the sensor chips employed for other interferometers like Mach Zehnder or were kept under a continuous flow of SSC-P (SSC 2.5X + 0.5% Young designs, characterized by Y-junctions and two Tween 20 + 10 mM CHAPS) at 10 μL min−1. Figure S2B 14660 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article Figure 1. BiMW sensor biofunctionalization for microRNA-21-5p detection. Representation of the three main steps in the development of a complementary hybridization assay for the detection of microRNA-21-5p. shows a sensorgram of the two-step reaction involved in the containing 0.72 μL of RT primer, 25 ng of total RNA and 0.5 covalent coupling of both, DNA probes and BSA, to the Si3N4 μL of MultiScribe reverse transcriptase, 0.75 μL of reverse sensor surface. transcription buffer (10×), 0.075 μL of dNTPs (100 mM), and Complementary Hybridization Assay Performance. 0.1 μL of RNase inhibitor according to a TaqMan MicroRNA Calibration curves in standard buffer conditions were obtained reverse transcription kit (Applied Biosystems, Massachusetts, by flowing different microRNA-21-5p solutions (from 0.5 to US) protocol. The cDNA mixture was incubated in PCR tubes 100 nM, 150 μL) dissolved in SSC 5X buffer (0.75 M in NaCl, for 16 °C 30 min, 42 °C 30 min, and 85 °C 5 min. qRT-PCR 0.075 M in sodium citrate) over the BiMW biofunctionalized was performed in a total volume of 10 μL using 5 μL of 5.5 biosensor surface at a 10 μL min−1 rate, using SSC 5X as TaqMan Fast Advanced Master Mix and 0.47 μL of primer for running buffer. Calibration curves were obtained by analyzing each microRNA in a ViiA7 (Applied Biosystems, Massachu- different microRNA-21-5p concentrations in triplicate. Cali- setts, US) PCR machine. The reactions were set up as follows: bration curves in plasma were generated by flowing different the initial step includes the UNG incubation at 50 °C for 2 min concentrations of microRNA-21-5p (ranging from 0.5 to 100 and polymerase activation at 95 °C for 20 s, followed by 40 nM) spiked in undiluted human plasma over the BiMW sensor cycles of 95 °C for 1 s (denature) and 60 °C (anneal/extend) surface at a flow rate of 10 μL min−1, using SSC-P (SSC 2.5X + for 20 s. The relative expression level was calculated using 0.5% Tween 20 + 10 mM CHAPS) as running buffer. In all 2−ΔΔCT (fold change).25 A median Ct across all samples was cases, DNA probe/microRNA interaction was disrupted by chosen as a calibrator and miR-16-5p as the endogenous injecting a 5 mM NaOH regeneration solution for 30 s at a control. Thermo Fisher microRNA Primer Assay ID: hsa-miR- constant flow rate, allowing the reuse of the sensor chips for 10 21-5p #000397; hsa-miR-16-5p #000391. cycles without affecting assay performance. Figure 1 shows a Data Analysis. The real-time sensorgrams were processed schematic representation of the complete biofunctionalization extracting the final response (Δφ) after signal stabilization protocol. once the whole sample volume has passed through the flow Clinical Sample Collection. A total of 20 patients from cell. Details of the fitting curves are described in the SI. the “Leon Daniello” Pneumophtisiology Clinical Hospital Cluj- Statistical analysis assessing the differences between healthy Napoca, Romania diagnosed with NSCLC were enrolled in our and LC groups was analyzed with the Mann−Whitney test study. Besides the LC patients, we enrolled 20 healthy subjects considering a p-value < 0.05 to be statistically significant. The in our study for blood sample donation (controls) (Table 2). correlation between the BiMW biosensor and qRT-PCR was analyzed by the Spearman test considering a p-value < 0.05. In Table 2. Lung Cancer Patients and Healthy Subjects order to evaluate the diagnosis capabilities of both the BiMW Included in the Study biosensor and qRT-PCR, receiver operating characteristic (ROC) curves were also performed. Statistical analysis to sample plasma type no. samples cancer stage origin discover any significant relationship between sensor response and cancer stages was done with the Kruskal−Wallis test, plasma 20 (50%) IIIA-(25%)/IIIB-(50%)/IV-(25%) LC patients (NSCLC) considering a p-value < 0.05. plasma 20 (50%) healthy subjects ■ RESULTS AND DISCUSSION Biosensor Assay and Analytical Characterization. In The plasma samples were isolated by centrifugation at 4200 order to demonstrate and evaluate the performance of the rpm/10 min at room temperature. From each patient blood, BiMW biosensor, an assay strategy based on direct hybrid- samples were collected according to the hospital protocol and ization was designed, employing a synthetic DNA probe fully the informed consent approved by the Ethical Committee no. complementary to the microRNA-21-5p as the capture probe, 438/2016. thus providing the required specificity. Figure 2 shows the RNA Extraction and qRT-PCR. Total RNA was extracted biosensor signals obtained for microRNA-21-5p detection. and isolated from plasma samples using a Plasma/Serum Slight signal fluctuations were observed, while the sample was Circulating and Exosomal RNA Purification Kit (Norgen entering the cell, attributed to the running buffer and the Biotek Corp., Ontario, Canada) according to the manufac- modulation method employed to convert the interferometric turer’s protocol. The RNA concentration was measured using a signal into a linear one (see Section 2.1 in the SI). However, NanoDrop-1000 spectrophotometer (ThermoFisher Scientific, this was corrected once all the sample had flown, resulting in Massachusetts, US). Total RNA concentration was ranged to stabilized signals. The sensor response (Δφ, rad) gradually 25 ng μL−1. microRNA-21-5p was selected for the plasma increased as the microRNA concentrations were higher. The microRNA profile analysis by qRT-PCR. The cDNA synthesis representation of the log−log variables adequately fitted to a was performed using 7.5 μL of reverse transcription mixture linear curve (see Figure 2B and Section 3 in the SI) for the 14661 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article Table 3. Interassay Variability for Buffer and Undiluted Plasma Calibration Curves (CC)a matrix parameters buffer A b LOD, pM CC1 −1.27 0.64 43 CC2 −1.18 0.59 20 mean ± SD −1.23 ± 0.06 0.61 ± 0.03 31.5 ± 16.2 %CV 5.3% 5.7% 51% plasma A b LOD, pM CC1 −0.96 0.56 18 CC2 −1.00 0.58 23 mean ± SD −0.98 ± 0.03 0.57 ± 0.01 20.5 ± 3.5 CV% 3.4% 2.6% 17% a Eq: log((Δφ) = A + b · log ([miRNA]) → Δφ = A·[miRNA]b. hybridization event. Plasma contains high amounts of proteins and other compounds that could generate nonspecific interactions or hinder the DNA probe−microRNA interaction (i.e., untreated biofunctionalized sensor chip resulted in extremely high signals of around Δφ ≈ 40 rad). Therefore, we combined the use of blocking agents (added to the surface to increase its biocompatibility and hydrophilicity) with Figure 2. BiMW-based microRNA-21-5p hybridization assay in mixtures of additives that help shield nonspecific interactions buffer. (A) Real-time sensorgrams showing the specific interaction (see Figure S3). According to the conditions tested, we of the DNA probe with different microRNA-21-5p concentrations. decided to employ a combination of BSA 20 mg mL−1 as the Nonspecific microRNA-210-3p (control) was measured at a blocking agent and Tween 20 0.5% + CHAPS 10 mM, concentration of 100 nM. (B) Calibration curve in hybridization nonionic, and zwitterionic surfactants, respectively, as the best buffer (SSC 5X) (log−log display). Each signal corresponds to the suitable additives that overall successfully removed all mean ± SD of triplicate measurements. IN (t ∼ 500 s) and OUT (t ∼ nonspecific interactions from human plasma (Δφ ≈ 0 rad). 1500 s) arrows indicate the start and end time of the injection, Pooled human plasma was spiked with different microRNA- respectively. 21-5p concentrations in the range from 0.5 to 100 nM (Figures 3 and S4 in the SI). The calibration curve performed in plasma range of microRNA concentrations analyzed (i.e., 0.5−100 showed an estimated LOD of 25 pM (R2 = 0.9701). The nM), being possible to establish an estimated LOD = 34 pM similar LOD value obtained compared with the one in (R2 = 0.9801). Even though there is not a consensus about the standard buffer conditions (34 pM) reveals that the biological concentration of microRNA in plasma due to the lack of fluid did not compromise the assay sensitivity, not affecting the standardization in RNA extraction and the variation between sensor surface or the DNA probe and its capabilities to quantification methodologies, some studies suggest that the hybridize. The assay in plasma showed a similar reproducibility microRNA concentration in human plasma might lie in the to the one observed in standard buffer conditions (see Table range of 105−108 copies mL−1 (i.e., fM−nM).26 According to 3) which confirms the suitability of the designed strategy for these LOD, the performance of our biosensor may provide the accurate detection of microRNA-21-5p with the BiMW enough analytical sensitivity for LC diagnosis. biosensor-based assay. Under these optimized conditions, the The assay specificity was also evaluated to guarantee the biosensor assay was further evaluated with real clinical samples. absence of nonspecific interactions of other microRNAs Clinical Validation of BiMW-Based microRNA-21-5p involved in LC (i.e., microRNA-210-3p). As we can observe Assay. We have assessed a set of 40 plasma clinical samples in Figure 2A, microRNA-210-3p (control) interacted neither from the Research Center for Functional Genomics, with the sensor surface nor with the DNA probe (i.e., sensor Biomedicine and Translational Medicine (Romania). The response Δφ = 0 rad after signal stabilization). Net sensor collection consisted of 20 LC plasma samples and 20 negative response confirms the absence of cross-reactivity and ensures samples from healthy donors. LC plasma samples were that the signals come exclusively from specific microRNA− collected from patients previously diagnosed with LC, complementary DNA probe interactions. specifically, NSCLC through bronchoscopy. All samples were The reproducibility of the assay was evaluated through the previously validated by qRT-PCR in the Research Center for interassay variability (replicates in different sensor chips). The Functional Genomics, Biomedicine and Translational Medi- CV% values obtained for buffer conditions were below the cine, with a positive result for microRNA-21-5p (Table S1). All maximum variability recommended for clinical analysis (15− the samples were analyzed with the BiMW biosensor, and a 20%)27 (slightly higher for the LOD) (Table 3), verifying the statistical comparison between healthy and LC individuals was good reproducibility and suitability of this hybridization assay. carried out. Figure 4A presents the distribution of the Human Plasma Effect on the Assay Performance. To biosensor response obtained by using our hybridization apply the described biosensor methodology for the analysis of assay, showing statistical significance differentiation between LC patients’ plasma samples, it is crucial to consider the healthy [median = 2.368] and LC patient [median = 3.563] influence of the plasma matrix on the sensor surface and the (p-value < 0.0001; p-value < 0.05). Additionally, an ROC curve 14662 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article of a diagnostic test to discriminate between healthy and patients by considering 1 as an excellent and 0.5 as random diagnosis. The BiMW AUC outcome was 0.87 (CI95%, 0.7616−0.9784), reflecting an appropriate diagnosis capability with a sensitivity of 80% and specificity of 80%. The BiMW biosensor performance was qualitatively compared to the standard technique qRT-PCR. Results are shown in Figure 4. As can be observed, qRT-PCR-based microRNA quantification was able to discriminate between healthy [median = 0.99] and LC patients [median = 3.7] in a statistically significant manner (p-value < 0.0001) (Figure 4B). ROC curves determined that, as the BiMW biosensor, the qRT-PCR methodology presented acceptable LC detection capabilities (AUC 0.8675 (CI95%, 0.75−0.985), sensitivity 75%, and specificity 95%) (Figure 4D). An additional correlation analysis was carried out by the Spearman test, showing a significant relationship between the quantification value obtained with qRT-PCR and the signal obtained with the BiMW biosensor for most of the samples, which might reflect the reliability of the BiMW biosensor assay (Spearman coefficient = 0.373, p-value = 0.018, and p-value < 0.05) (data not shown). These results corroborate the competitive performance of our BiMW technology compared with the benchmarked qRT-PCR. Moreover, the BiMW biosensor can provide highly accurate detection of microRNA in less than 45 Figure 3. BiMW-based microRNA-21-5p hybridization assay in min, with diagnostic reliability equivalent to qRT-PCR, plasma. (A) Real-time sensorgrams showing the specific interaction without previous amplification or purification steps, revealing of the DNA probe with different microRNA-21-5p concentrations spiked in human plasma. (B) Calibration curve in human plasma the potential of the nanophotonic waveguide interferometer (log−log display). Each signal corresponds to the mean ± SD of biosensor technology for LC clinical diagnosis. triplicate measurements. IN (t ∼ 500 s) and OUT (t ∼ 3000 s) Finally, to test the capabilities of the BiMW biosensor, a arrows indicate the start and end time of the injection, respectively. study with the 20 clinical samples from LC patients was carried out to assess a possible correlation between the cancer stage was performed to evaluate the diagnostic specificity and and the levels of microRNA-21-5p in the plasma. To stratify sensitivity of the BiMW biosensor technology (Figure 4C). patients according to their oncological status, the TNM The area under curve (AUC) value determines the capability classification was used in which the size and extent of the Figure 4. microRNA-21-5p detection in clinical samples based on (A) BiMW sensor and (B) qRT-PCR in 20 healthy and 20 LC individuals. Mann−Whitney test p-value < 0.0001. Outliers are also shown. ROC curve analysis of (C) BiMW technology and (D) qRT-PCR. AUC, sensitivity, and specificity values are reported. 14663 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article primary tumor (T), the number of nearby lymph node specificity of 80% in both cases. We have also conducted a invasion (N), and the absence or presence of metastasis (M) small-scale study to assess the capabilities of our strategy for were analyzed (Table S1 in the Supporting Information). We early diagnosis of LC, although clinical samples at stages 0, I, evaluated 20 plasma samples from different LC stages [IIIA (n or II would be necessary to carry out a complete analysis and = 5), IIIB (n = 10), and IV (n = 5)] (Figure 5). to extract relevant conclusions. The presented work represents a relevant step toward the implementation of this biosensor for clinical diagnosis. The obtained results place our biosensor device as an accurate and reliable tool for rapid and direct detection of microRNA, becoming a potential alternative tool for early cancer diagnosis and treatment in a noninvasive manner and with great perspectives in clinical practice. ■ * ASSOCIATED CONTENT sı Supporting Information The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acs.analchem.2c02895. Description of the BiMW biosensor device, and details Figure 5. Correlation cancer stage vs microRNA-21-5p concentration. Sensor signal for the 20 LC plasma samples from individuals in of reagents and buffers, biosensor chip biofunctionaliza- different oncological stages (IIIA, IIIB, and IV). Kruskal−Wallis test tion, data analysis, and additional results for the (p-value = 0.4833; p-value > 0.05). Outliers are also shown. validation of the methodology (PDF) As Zhang et al. reported,28 the overexpression of microRNA- 21-5p, and therefore its concentration in body fluids, increases ■ AUTHOR INFORMATION Corresponding Authors depending on the TNM stage, being higher in advanced TNM Maria Carmen Estévez − Nanobiosensors and Bioanalytical cancers. Nevertheless, Figure 5 reveals that despite the Applications Group (NanoB2A), Catalan Institute of capability of the BiMW to diagnose LC, no conclusive Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- evidence could be extracted regarding a possible correlation BBN and BIST, 08193 Bellaterra, Barcelona, Spain; between the LC stage and the epigenetic response reflected in orcid.org/0000-0003-3694-7186; the microRNA-21-5p concentration in plasma. Statistical Email: mcarmen.estevez@icn2.cat analysis shows a nonsignificant difference in the microRNA Laura M. Lechuga − Nanobiosensors and Bioanalytical concentration between average groups (IIIA [4.290]; IIIB Applications Group (NanoB2A), Catalan Institute of [4.383]; IV [4.074], with a p-value = 0.4833). The limited Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- number of samples and above all the difficulty in collecting BBN and BIST, 08193 Bellaterra, Barcelona, Spain; early-stage LC samples, since all LC samples presented orcid.org/0000-0001-5187-5358; Email: laura.lechuga@ advanced TNM stage LC (III and IV), hinder the capability icn2.cat to prognosticate and prompt diagnose this type of cancer with our biosensor. We acknowledge the necessity of completing Authors this study with an extended number of samples and a more Olalla Calvo-Lozano − Nanobiosensors and Bioanalytical variety of LC samples (i.e., from stage 0 to stage IV). Despite Applications Group (NanoB2A), Catalan Institute of the mentioned limitations, the BiMW technology exemplifies Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- the benefit that a quantitative microRNA biosensor assay can BBN and BIST, 08193 Bellaterra, Barcelona, Spain; provide for monitoring cancer progression. orcid.org/0000-0001-5486-2237 Pablo García-Aparicio − Nanobiosensors and Bioanalytical ■ CONCLUSIONS We have demonstrated a novel BiMW biosensor methodology Applications Group (NanoB2A), Catalan Institute of Nanoscience and Nanotechnology (ICN2), CSIC, CIBER- for the fast, direct, and quantitative identification of the LC BBN and BIST, 08193 Bellaterra, Barcelona, Spain -related microRNA-21-5p biomarker in human plasma. Our Lajos-Zsolt Raduly − Research Center for Functional advanced biosensor, based on a nanowaveguide interferometric Genomics, Biomedicine and Translational Medicine, technology, offers high sensitivity and label-free analysis in real University of Medicine and Pharmacy “Iuliu Hatieganu”, time. The complementary hybridization strategy consisting of 400337 Cluj-Napoca, Romania covalent immobilization of DNA probes that provides the Ioana Berindan-Neagoe − Research Center for Functional assay specificity reached excellent LOD in plasma (pM range) Genomics, Biomedicine and Translational Medicine, enabling one-step detection and quantification of microRNA- University of Medicine and Pharmacy “Iuliu Hatieganu”, 21-5p in LC samples. A clinical validation with healthy and LC 400337 Cluj-Napoca, Romania; orcid.org/0000-0001- patients (n = 40) showed excellent discrimination between 5828-1325 healthy and cancer samples (p-value < 0.0001), with a Manuela Ferracin − Department of Experimental, Diagnostic performance similar to the one of established techniques and Specialty Medicine (DIMES), University of Bologna, such as qRT-PCR and with the additional advantage of 40126 Bologna, Italy avoiding RNA extraction or amplification steps. In addition, Complete contact information is available at: the clinical validation demonstrated diagnostic sensitivity and https://pubs.acs.org/10.1021/acs.analchem.2c02895 14664 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665 Analytical Chemistry pubs.acs.org/ac Article Author Contributions Cohen, D.; Chajut, A.; Barshack, I. Nat. Biotechnol. 2008, 26, 462− O.C.-L. carried out the optimization of the assay and the 469. BiMW measurements. P.G.-A. collaborated on the optimiza- (11) Pencheva, N.; Tavazoie, S. F. Nat. Cell Biol. 2013, 15, 546−554. tion of the complementary hybridization assay. I.B.-N. and (12) Ouyang, T.; Liu, Z.; Han, Z.; Ge, Q. Anal. Chem. 2019, 91, 3179−3186. L.Z.-R. collected, treated, and validated the plasma clinical (13) Huertas, C. S.; Calvo-Lozano, O.; Mitchell, A.; Lechuga, L. M. samples through qRT-PCR. O.C.-L. conceptualized the Front. Chem. 2019, 7, 724. experiments and wrote the presented manuscript. M.C.E., (14) Schmieder, S.; Weißpflog, J.; Danz, N.; Klotzbach, U.; Sonntag, I.B.-N., M.F., and L.M.L. took part in the discussion of the F. Curr. Dir. Biomed. Eng. 2016, 2, 135−138. following manuscript and revised it critically for important (15) Liu, R.; Wang, Q.; Li, Q.; Yang, X.; Wang, K.; Nie, W. Biosens. intellectual content. All authors have read and agreed to the Bioelectron. 2017, 87, 433−438. published version of the manuscript. (16) Li, X.; Cheng, W.; Li, D.; Wu, J.; Ding, X.; Cheng, Q.; Ding, S. Biosens. Bioelectron. 2016, 80, 98−104. Notes (17) Graybill, R. M.; Cardenosa-Rubio, M. C.; Yang, H.; Johnson, The authors declare no competing financial interest. M. D.; Bailey, R. C. Anal. Methods 2018, 10, 1618−1623. ■ ACKNOWLEDGMENTS The ICN2 acknowledges the financial support from the (18) Liu, Q.; Shin, Y.; Kee, J. S.; Kim, K. W.; Mohamed Rafei, S. R.; Perera, A. P.; Tu, X.; Lo, G. Q.; Ricci, E.; Colombel, M.; Chiong, E.; Thiery, J. P.; Park, M. K. Biosens. Bioelectron. 2015, 71, 365−372. Spanish Research Agency (AEI), Ministry of Science, (19) Zinoviev, K. E.; González-Guerrero, A. B.; Domínguez, C.; Lechuga, L. M. J. Light. Technol. 2011, 29, 1926−1930. Innovation, and Universities through the PREDICT (20) Gavela, A. F.; García, D. G.; Ramirez, J. C.; Lechuga, L. M. (TEC2016-78515-R) and EROICA PID2019-105132RB-I00) Sensors 2016, 16, 285. projects. ICN2 is funded by the CERCA Program/Generalitat (21) Maldonado, J.; Estévez, M. C.; Fernández-Gavela, A.; González- de Catalunya and supported by the Severo Ochoa Centres of López, J. J.; González-Guerrero, A. B.; Lechuga, L. M. Analyst 2020, Excellence program funded by the Spanish Research Agency 145, 497−506. (AEI, grant no. SEV-2017-0706). O.C-L. acknowledges the (22) Maldonado, J.; González-Guerrero, A. B.; Fernández-Gavela, economic support from the Spanish Ministry of Science and A.; González-López, J. J.; Lechuga, L. M. Diagnostics 2020, 10, 845. Innovation and the Spanish Research Agency and the (23) Huertas, C. S.; Fariña, D.; Lechuga, L. M. ACS Sensors 2016, 1, European Social Fund (ESF) (ref. BES-2017-080527) linked 748−756. to the PREDICT project. Research Center for Functional (24) González-Guerrero, A. B.; Maldonado, J.; Dante, S.; Grajales, D.; Lechuga, L. M. J. Biophotonics 2017, 10, 61−67. Genomics, Biomedicine and Translational Medicine acknowl- (25) Livak, K. J.; Schmittgen, T. D. Methods 2001, 25, 402−408. edges “Clinical and economic impact of personalized targeted (26) Mitchell, P. S.; Parkin, R. K.; Kroh, E. M.; Fritz, B. R.; Wyman, anti-microRNA therapies in reconverting lung cancer chemo- S. K.; Pogosova-Agadjanyan, E. L.; Peterson, A.; Noteboom, J.; resistance”, CANTEMIR, no. POC-P_37_796/2016. This O’Briant, K. C.; Allen, A.; Lin, D. W.; Urban, N.; Drescher, C. W.; work was supported by the FLAG-ERA grant LEGOCHIP, Knudsen, B. S.; Stirewalt, D. L.; Gentleman, R.; Vessella, R. L.; funded by the Italian National Research Council and the Nelson, P. S.; Martin, D. B.; Tewari, M. Proc. Natl. Acad. Sci. U. S. A. Spanish National Research Agency (PCI2019-111890-2). 2008, 105, 10513−10518. (27) USFDA. Bioanalytical Method Validation, Guidance for Industry; ■ REFERENCES (1) Cancer Today - International Agency for Research on Cancer US Food and Drug Administration, 2018. (28) Zhang, J. G.; Wang, J. J.; Zhao, F.; Liu, Q.; Jiang, K.; Yang, G. H. Clin. Chim. Acta 2010, 411, 846−852. https://gco.iarc.fr/today/home (accessed Jan 12, 2022), DOI: DOI: 10.1097/ADM.0000000000001074. (2) Siegel, R. L.; Miller, K. D.; Jemal, A. CA Cancer J. Clin. 2020, 70, 7−30. (3) Latimer, K. M.; Mott, T. F. Am. Fam. Physician 2015, 91, 250− 256. (4) Cryer, A. M.; Thorley, A. J. Pharmacol. Ther. 2019, 198, 189− 205. (5) Usman, F.; Dennis, J. O.; Aljameel, A. I.; Ali, M. K. M.; Aldaghri, O.; Ibnaouf, K. H.; Zango, Z. U.; Beygisangchin, M.; Alsadig, A.; Meriaudeau, F.; Sa, A. I. A. Chemosensors 2021, 9, 326. (6) Molina, R.; Marrades, R. M.; Augé, J. M.; Escudero, J. M.; Viñolas, N.; Reguart, N.; Ramirez, J.; Filella, X.; Molins, L.; Agustí, A. Am. J. Respir. Crit. Care Med. 2016, 193, 427−437. (7) Diaz-Lagares, A.; Mendez-Gonzalez, J.; Hervas, D.; Saigi, M.; Pajares, M. J.; Garcia, D.; Crujerias, A. B.; Pio, R.; Montuenga, L. M.; Zulueta, J.; Nadal, E.; Rosell, A.; Esteller, M.; Sandoval, J. Clin. Cancer Res. 2016, 22, 3361−3371. (8) Lu, H. Y.; Qin, J.; Han, N.; Lei, L.; Xie, F.; Li, C. Onco. Targets. Ther. 2018, 11, 2217. (9) Iqbal, M. A.; Arora, S.; Prakasam, G.; Calin, G. A.; Syed, M. A. Mol. Aspects Med. 2019, 70, 3−20. (10) Rosenfeld, N.; Aharonov, R.; Meiri, E.; Rosenwald, S.; Spector, Y.; Zepeniuk, M.; Benjamin, H.; Shabes, N.; Tabak, S.; Levy, A.; Lebanony, D.; Goren, Y.; Silberschein, E.; Targan, N.; Ben-Ari, A.; Gilad, S.; Sion-Vardy, N.; Tobar, A.; Feinmesser, M.; Kharenko, O.; Nativ, O.; Nass, D.; Perelman, M.; Yosepovich, A.; Shalmon, B.; Polak-Charcon, S.; Fridman, E.; Avniel, A.; Bentwich, I.; Bentwich, Z.; 14665 https://doi.org/10.1021/acs.analchem.2c02895 Anal. Chem. 2022, 94, 14659−14665