Letter Cite This: ACS Sens. 2017, 2, 1761−1766
pubs.acs.org/acssensors
Tracking Silent Hypersensitivity Reactions to Asparaginase during Leukemia Therapy Using Single-Chip Indirect Plasmonic and Fluorescence Immunosensing David M. Charbonneau,†,‡ Julien Breault-Turcot,†,§ Daniel Sinnett,∥,⊥ Maja Krajinovic,∥ Jean-Marie Leclerc,∥ Jean-François Masson,†,§ and Joelle N. Pelletier*,†,‡,#,¶ †
Département de chimie, ⊥Département de pédiatrie, and ¶Département de biochimie, Université de Montréal, Montréal, Québec H3T 1J4, Canada ‡ PROTEO Network, Université Laval, Québec, Québec G1V 0A6, Canada § Centre Québécois sur les Matériaux Fonctionnels (CQMF), Université de Sherbrooke, Québec, Québec J1K 2R1, Canada ∥ Centre de recherche, CHU Sainte-Justine, Montréal, Québec H3T 1C5, Canada # Center for Green Chemistry and Catalysis (CGCC), Montréal, Québec H3A 0B8, Canada S Supporting Information *
ABSTRACT: Microbial asparaginase is an essential component of chemotherapy for the treatment of childhood acute lymphoblastic leukemia (cALL). Silent hypersensitivity reactions to this microbial enzyme need to be monitored accurately during treatment to avoid adverse effects of the drug and its silent inactivation. Here, we present a dual-response antiasparaginase sensor that combines indirect SPR and fluorescence on a single chip to perform ELISA-type immunosensing, and correlate measurements with classical ELISA. Analysis of serum samples from children undergoing cALL therapy revealed a clear correlation between single-chip indirect SPR/fluorescence immunosensing and ELISA used in clinical settings (R2 > 0.9). We also report that the portable SPR/fluorescence system had a better sensitivity than classical ELISA to detect antibodies in clinical samples with low antigenicity. This work demonstrates the reliability of dual sensing for monitoring clinically relevant antibody titers in clinical serum samples. KEYWORDS: Dual SPR/fluorescence sensing, ELISA, L-asparaginase, immunosensing, portable sensor
M
cules.4,5 Commercial sensing instruments such as the Biacore (GE Healthcare) offer an alternative to classical ELISA based on immunosorbing, and thus, they have been applied for monitoring markers and molecules in clinical samples for a series of diseases.6 Despite the fact that SPR sensing is perceived as a potential POCT technology, commercial instruments are not adapted for POCT because of their cost, size, and complexity. Smaller and portable SPR sensors have thus been developed in the past decade to facilitate application of SPR sensors for monitoring markers near the patient.7 Furthermore, plasmonic sensing in complex biological samples
icrobial asparaginase is an essential component of chemotherapy for childhood acute lymphoblastic leukemia (cALL). However, clinical application of Escherichia coli asparaginase II (EcAII) is complicated by the frequent development of allergy and silent hypersensitivity reactions.1,2 Silent immunological responses to EcAII are associated with the development of antibodies that may interfere with the efficacy of the treatment. The detection of anti-EcAII antibodies (mainly IgG) has been reported to have a prognostic significance for treatment efficacy.3 The main method for measuring antibodies associated with silent immunological responses, ELISA, is not ideally suited for point-of-care testing (POCT) as it is lengthy and multistep. Surface plasmon resonance (SPR) sensing has been increasingly applied for monitoring and quantifying biomole© 2017 American Chemical Society
Received: August 17, 2017 Accepted: November 23, 2017 Published: November 23, 2017 1761
DOI: 10.1021/acssensors.7b00584 ACS Sens. 2017, 2, 1761−1766
Letter
ACS Sensors
focused on an avalanche photodiode. We report an improved correlation with classical ELISA that we performed in parallel, according to the standard method used in clinical settings for detecting silent hypersensitivity reactions to EcAII. The standard immunosorbing method involves indirect detection through the use of a secondary antibody that is HRPconjugated to detect conversion of o-phenylenediamine dihydrochloride (OPD) by absorbance at 490 nm, and is referred to as “ELISA absorption detection” hereafter.14 EcAII is the main asparaginase preparation used in front-line cALL therapy. As it originates from E. coli, it is a potential immunogen. EcAII is used in its native form (Elspar, Kidrolase, and L-asparaginase Medac) or in a PEGylated form designed to reduce immune response (Oncaspar and Calspargase Pegol). In the case where allergy to EcAII is observed, the asparaginase preparation is changed to ErAII (Erwinase) from Erwinia chrysanthemi which exhibits low cross-reactivity.15 A recent study from St. Jude Children’s Research Hospital reported an occurrence of 41% (n = 410) clinical allergy to native Elspar. Out of these 169 patients with overt allergy, antibodies were detected in 147 patients (87%). However, among 241 patients (59%) who did not display symptoms of clinical allergy, 89 patients (37%) had detectable anti-EcAII antibodies. Such silent immunological response may result in silent inactivation by neutralizing antibodies, directly interfering with enzyme activity and/or by binding antibodies leading to antibody opsonization and drug clearance.2,16 Such immune response adds to the metabolic burden of patients and should be rapidly detected to halt administration of the allergen. Surprisingly, a study from the Dana Farber Cancer Institute showed that 12% (n = 232) of children treated with the PEGylated Oncaspar also developed clinical allergy, compared to 9% (n = 231) for Kidrolase.17 Furthermore, the Dutch Childhood Oncology Group has reported that 22% (n = 91) of children treated with the native Medac during the induction phase of cALL treatment developed clinical allergy to the PEGylated Oncaspar during the post-induction phase, while 8% showed silent inactivation.18 Overall, silent hypersensitivity occurs in 5−46% of the children treated with native and PEGylated EcAII.15,19−21 This highlights the need for rapid and accurate detection of anti-EcAII antibodies in a timely manner to prevent potential future allergic reactions and improve the global benefit of the chemotherapy treatment. At the outset of this study, 87 serum samples obtained from 18 children undergoing cALL chemotherapy at CHU SainteJustine (9 males aged from 3 to 15 years old and 9 females aged from 2 to 11 years old) were analyzed by standard ELISA absorption detection. The samples were divided into 5 groups according to whether the patients were being actively treated or had completed treatment, and whether they received PEGylated or non-PEGylated EcAII. Table S1 provides details about the treatments that were administered along with the associated risk factor, observation of clinical allergy and toxicity. Serum samples from 10 healthy children (3 males aged from 15 to 17 years old and 7 females aged 2−17 years old) who never received asparaginase served as the reference (negative controls), Table S2. The ELISA absorption detection was conducted for total anti-EcAII IgG antibodies according to a protocol adapted from Wang and coauthors14 (Figure S1; complete data set is shown in Figure S2, subset shown in Figure 2). Samples were determined as positive or negative according to thresholds determined as defined in methods (Supporting Information).
such as serum or plasma required improved sensor surfaces to limit surface fouling and detection strategies which are increasingly available.6,8 The combination of smaller portable SPR instruments and better surface chemistry/sensing strategy is thus promising for the use of SPR in clinical settings. To address the challenges associated with POCT and plasmonic sensing in serum, we recently reported a highsensing and low-fouling anti-asparaginase sensor chip for use in a portable SPR sensing device.9 We demonstrated the necessity to monitor an indirect SPR signal using a secondary antibody for accurate detection of anti-EcAII antibodies in undiluted clinical serum because sample-to-sample variation in the bulk medium affected the reliability of direct plasmonic sensing.10,11 Dual SPR/fluorescence detection should reduce the incidence of false positives/negatives because positive signals must be generated in both channels and be in quantitative agreement for a result to be valid. It will also increase the sensitivity and specificity of the sensor because the methods are not sensitive to the same interference. We also demonstrated that fluorescence detection can be more sensitive than SPR sensing, therefore extending the dynamic range of a dual sensor.12 Thus, the use of a dual readout using SPR and on-chip fluorescence immunosensing is attractive for improving the confidence in results. Here, we report an optimized secondary SPR sensor with parallel fluorescence detection. As a proof-of-concept experiment, we tested our portable dual SPR-fluorescence immunosensing device by monitoring anti-asparaginase antibodies in serum from children undergoing cALL chemotherapy. This device was built from a standard P4-SPR unit (Affinité Instruments) above which a specifically designed fluorescent microscope was mounted (Figure 1). A broadband white light
Figure 1. Configuration of the dual SPR-fluorescence device, described in ref 12. Additional details on the optical system are provided in the Supporting Information.
LED illuminates the gold mirror on the SPR chip in total internal reflection mode and is collected with optical fibers for wavelength interrogation, which is is a robust approach to miniaturization of SPR sensing devices.13 This monitors the shift of the resonance wavelength as a function of the analytes bound to the gold surface. Measurement of the fluorescence generated by an HRP-conjugated secondary antibody (in this case, or a source of fluorescence in general) occurs on the same chip: a fluorescence epi-microscope with a 532 nm fibercoupled laser source is focused in the solution of the P4-SPR fluidic cell. The emitted photons are collected, directed through a long-pass 532 nm emission filter to reduce background, and 1762
DOI: 10.1021/acssensors.7b00584 ACS Sens. 2017, 2, 1761−1766
Letter
ACS Sensors
In parallel with the ELISA absorption detection assays, we applied our recently reported bioreceptor for detecting antiEcAII antibodies in serum using surface plasmon resonance (SPR). The sensor biochip consists of a gold-coated glass prism modified with a low-fouling peptide to which EcAII is immobilized with a high surface density.9 We previously demonstrated the necessity of monitoring a secondary antibody to counteract the high sample-to-sample variation in refractive index of serum samples, for accurate sensing of anti-EcAII antibodies in undiluted serum.10 Use of the anti-human HRP-conjugated IgG secondary antibody to monitor a secondary SPR signal is conceptually comparable to performing an ELISA assay on the SPR sensor chip, except that the binding events are monitored in real-time. Binding can additionally be monitored in parallel by detecting the fluorescence associated with the HRP-catalyzed conversion of Ampliflu Red into resorufin using our adapted dual SPR/ fluorescence instrument.12 Thus, an HRP-conjugated secondary antibody both procures a secondary SPR signal and catalyzes the formation of a fluorescent product; this will be referred to as SPR and ELISA fluorescence detection hereafter. We thus performed dual indirect SPR and ELISA fluorescence detection of anti-EcAII in undiluted sera from two positive samples (PO01-S3, PO01-S4) as well as three negative samples (PO01-S5, PO02-S1, and PSC01-S2), as determined by ELISA absorption detection. Optimized secondary detection showed an excellent correlation (R2 = 0.9721) between indirect SPR and fluorescence measurements on the same chip for each sample (Figure 3). Most importantly, a good correlation was observed between indirect ELISA absorption detection and both indirect SPR (R2 = 0.9473) and ELISA fluorescence detection (R2 = 0.9301) measurements performed on the gold chip (Figure 4A). The strong correlation observed between either SPR or fluorescence (using undiluted sera) with ELISA absorption detection at 1:400 serum dilution
Figure 2. Anti-asparaginase antibodies monitored by indirect ELISA absorption detection in clinical sera. The assay was performed at serum dilution of 1:400 (1:10 000 sAb/0.01% H2O2). Negative controls are shown in pink and the pooled negative control sample (equal mixture) is in dark pink. Serum samples from treated patients (15 out of 87) are shown in gray. The threshold for determination of silent hypersensitivity is defined at 2.58 standard deviations above the mean value of the pooled negative controls (pink dashed line).2,14 Results from two triplicate experiments are shown (n = 6), p-value
