Reliable and sensitive detection of pancreatic cancer marker by gold nanoflower-based SERS mapping immunoassay

Highlights

• Uniformly distributed Au-nanoflowers on Si-wafer as support for capture surface.

• LSPR coupling between Au-nanoflowers helps improve detection sensitivity.

• SERS mapping reduces spot-to-spot variation in quantitative SERS.

• Incubating capture surface in microwell plates significantly reduces positive error.

Abstract

Reliability of a method for cancer detection is a key issue. It is because of fear for false positive and false negative results that no tumor marker has been recommended for cancer screening test yet. Surface-enhanced Raman scattering (SERS) is highly sensitive and a promising technique for cancer detection. However, huge spot-to-spot variation in SERS signal and false positive result in SERS-based immunoassay degrade its reliability. In this work, we designed gold nanoflower (AuNF)-based SERS method for detection of pancreatic cancer marker MUC4 and demonstrated approaches to improve reliability of the method. We employed Raman mapping over a large area of the substrate to minimize effect of spot-to-spot variation. Furthermore, the use of microwell plate for incubation of capture substrate minimized blank value significantly. The designed method is sensitive enough to detect MUC4 down to 0.1 ng mL⁻¹. The use of AuNFs to construct both ERL and capture surface has contributed to the improved detection sensitivity.

Introduction

The International Agency for Research on Cancer estimated 18.1 million new cases of cancer and 9.6 million deaths from cancer in 2018, indicating that cancer remained a major public health problem regardless of decades of efforts.[1], [2] Cancer is asymptomatic at early stage and most patients are diagnosed with advanced cancer that is difficult for treatment. Hence, in parallel to the prevention and treatment strategies, there is a huge demand to develop reliable early detection techniques. Recently, SERS has become a promising method for cancer detection as it has ultrahigh detection sensitivity, specificity and little or no sample preparation.[3], [4], [5], [6], [7], [8], [9] However, there are issues that need to be addressed to improve reliability of quantitative SERS, and realize its practical application for cancer detection.

Only less than 10% of pancreatic cancer patients can survive for five years, mainly due to lack of early detection and effective treatments.[10] To date, both chemotherapy and radiation therapy have been found ineffective in treatment of pancreatic cancer,[11] and hence surgery is the only curative and palliative treatment. Unfortunately, most patients are diagnosed with advanced pancreatic cancer that cannot be treated by resection.[12] Thus, there is a need for early detection of pancreatic cancer. Early detection could be made possible by monitoring level of biomarkers in the serum. Lack of specific tumor markers remains a challenging issue in serum-based detection of cancer.[13], [14] Recently, MUC4 (a transmembrane glycoprotein) has been found to aberrantly overexpressed in most pancreatic cancers, but not detected in chronic pancreatitis and in the normal pancreas.[11], [15] This makes MUC4 a potential pancreatic cancer marker for early detection.[16] However, the serum level of MU4 is below the detection limits of immunoassay platforms routinely used in clinics. In fact, detection of MUC4 in the serum was made possible for the first time by making use of SERS-based immunoassay.[17]

SERS has ultra-high detection sensitivity down to a single molecule and there are many more approaches being proposed by researchers in the field to further improve detection sensitivity.[18], [19], [20] Unfortunately, spot-to-spot variation in SERS signal remained a serious inherent challenge that degrades reliability and hinders practical applicability of quantitative SERS.[21] The challenge is even worse in SERS-based sandwich immunoassay as there are a lot more heterogeneous processes involved.[3] Non-uniform immobilization of capture antibody on a support surface, heterogeneous binding of antigen and extrinsic Raman label (ERL) on the capture surface are among the issues. Even in label free direct detection, such variability becomes significant as the concentration of analyte decreases. Designing uniform SERS substrate is the common approach to address this issue though this alone cannot eliminate the challenge. SERS measurement in homogeneous solution phase is another approach. [22] Using spinning sample stage could also be an option. It is also good idea to use objective lens with larger laser spot size for acquisition of SERS signals from a larger area of a substrate per measurement. In this work, we employed Raman mapping over a large area of substrate to minimize the effect of spot-to-spot variation in SERS signal.

High blank value, which is indicative of false positive result, is another serious challenge related to SERS-based sandwich immunoassay. The assumption in sandwich immunoassay is that ERLs that remain on capture surface and detected by final measurement are only those that are covalently bonded to captured antigen. On the other hand, ERL can also be physically adsorbed and remain on capture surface though it is not bonded to any antigen. As a result, after long incubation time, even blank sample may produce significant SERS signal. Blocking non-specific binding cites and adequate washing procedures are the common measures employed to minimize such error. The error becomes even worse if the ERL partially dries on capture surface. In this work, microwell plates were employed for incubation of capture surface to minimize blank value. By adopting these approaches, the overall reliability of the measurement can be significantly enhanced and make SERS-based analysis a viable solution for clinical application of precise cancer detection.