Advancements in SERS‐based biological detection and its application and perspectives in pancreatic cancer

Surface‐enhanced Raman spectroscopy (SERS) has become an essential bio‐detection technique. Due to its high sensitivity, good signal specificity, and resistance to photobleaching, SERS has been widely used in biomedical research fields such as molecular imaging, tumor diagnosis, and drug monitoring. This review focuses on the progress of SERS in biomedical applications. We first introduce the basic principle of SERS and the progress of substrate research. Then, we summarize the latest research progress on SERS in drug monitoring, cell and exosome detection, tumor imaging, and detection platforms combining microfluidic and lateral flow technologies. Subsequently, the applied research of SERS in early diagnosis of pancreatic cancer and drug efficacy monitoring is described. Finally, the future development direction and possible challenges of SERS in tumor diagnosis and treatment are proposed.


INTRODUCTION
Pancreatic cancer (PC) has an extremely high mortality rate, with an overall 5-year survival rate is only 10%. 1,2urgery in the early stages of PC is the most effective way to improve patient survival.However, approximately 80% of patients are already at an advanced stage when first diagnosed.In such cases, patients cannot receive surgical treatment, while palliative therapy becomes the only option, which is the main reason for reduced survival. 3,4herefore, increasing the early PC diagnosis rate is crucial to improving patient prognosis.The main diagnostic methods for PC are imaging diagnosis, histological and cytological biopsy for pathology diagnosis, serum tumor marker testing, genetic diagnosis, and circulating tumor cell testing. 5Although imaging diagnosis is compassionate and specific in the advanced stages of PC, it is challenging to accurately diagnose the early stages of PC.In addition, the endoscopic ultrasound-guided fine-needle aspiration cytological pathology is a more precise method. 6Still, it is invasive, costly, high risk, and therefore unsuitable for early PC diagnosis.Genetic and tumor marker testing provides a non-invasive way for early PC diagnosis. 7,8owever, genetic and tumor marker tests are mainly based on polymerase chain reaction (PCR) 9 and enzymelinked immunosorbent assay (ELISA). 10Although PCR and ELISA are susceptible and specific methods, they are expensive, time consuming, and require trained personnel and restrictive experimental conditions. 11Surfaceenhanced Raman spectroscopy (SERS) is increasingly acknowledged as a non-invasive and highly sensitive approach for tumor diagnosis. 12,13Current research indicates that using both label-free and labeled SERS probes, the detection of exosomes and biomarkers associated with PC can facilitate early and precise diagnosis. 14The labeled SERS assay is constructed on the foundation of noble metal nanoparticles (NPs) and monoclonal antibodies, thereby establishing a SERS immune array.Label-free SERS probes have the potential to co-target and identify exosomes and biomarkers, directly detecting PC cells.In the SERS pro-cess, original Raman signals of PC cells are acquired, and characteristic Raman peaks are identified through machine learning to create a predictive model, enabling the accurate distinction of PC cells from normal cells and other tumor cells. 15

SERS PRINCIPLE AND MECHANISM
SERS is an emerging analytical method that enables the detection of single molecular weight samples and provides rich fingerprint information of the molecular structure. 16,179][20] Among these mechanisms, EM plays a major role when there is a local surface plasmon resonance (LSPR) excited by excitation light or a strong local electromagnetic field enhancement induced by Mie scattering resonance (Figure 1A). 21,22The region of EM between NPs is referred to as the "hot spot."The proximity between two NPs is directly proportional to the prominence of the hot spot effect and the subsequent amplification of the Raman signal (Figure 1B).][25] The CM is mainly due to intermolecular charge transfer.Efficient charge transfer is favored when the highest occupied molecular orbital and lowest unoccupied molecular orbital of the molecule coincide with the valence conduction band energy level of the substrate.This leads to a change in the polarization tensor and electron density distribution of the molecule, resulting in the SERS effect (Figure 1C,D). 26Compared to the EM mechanism, it has a much shorter scope effect on the Åm range, requiring direct adsorption or chemical binding of the reporter to the metal surface.8][29] Furthermore, EM and CM coexist, with CM usually contributing a minor portion of the total enhancement.Particularly in semiconductors, the SERS enhancement mechanism frequently emerges from the interplay of various mechanisms, such as molecular resonance, charge-transfer resonance, electromagnetic resonance, and Mie resonance. 21,30

SERS BIOLOGICAL DETECTING SUBSTRATE
The SERS effect is closely related to the substrate, and gold and silver NPs are widely used as SERS-active substrates because of their excellent enhancement effect on numerous molecules, simple fabrication method, and convenience of storage.[36]

Noble metal SERS substrates
It has been found that noble metal NPs such as gold and silver exhibit excellent LSPR properties in the visiblenear infrared region, as their application as SERS optical substrates has attracted much attention.Wang et al. 37 prepared silver NP aggregates of different morphologies by hydrothermal method.Silver NPs with varying aggregation can be produced using anodic aluminum oxide as the experimental substrate (Figure 2A).It was used as a SERS substrate for the detection of pyridine molecules.The EF of the pyridine molecule was found to be 10 7 at a concentration of 10 −10 M, which proved that it has a good SERS enhancement effect.Although silver was the first SERSactive material discovered, its biological toxicity limits the application of silver-based SERS NPs in vivo.However, gold nanoparticles (GNPs) with excellent biostability and safety are currently the most commonly used SERS tagging cores.Sun and coworkers 38 prepared Au nanodisk arrays on Si substrates by combining X-ray printing with electronbeam vapor deposition (Figure 2B).They controlled the reaction conditions to effectively regulate the nanodisk size and distribution density.The results showed that the Au nanodisk arrays were used as SERS-active substrates to detect R6G probe molecules with the lowest detection limit of 10 −8 M and the SERS EF of 10 6 , demonstrating excellent SERS activity.The average EF of Pt and Pd substrates is 1.4 × 10 5 , two orders of magnitude lower than that of Au/Ag and Si substrates.The detection limit of Pt/Pd substrates is also 100 times worse than that of the previously discussed Au, Ag, and Si groups, which is about 2.9 × 10 −9 M. [39][40][41] The enhancement effect of SERS in NPs, driven by the manipulation of the composition, size, morphology, spacing, and surrounding media of noble metal NPs, has been significantly augmented due to recent diverse research initiatives.Notably, nanostructures with internal nanogaps embedded with Raman molecules have shown promising potential for biomedical applications.SERS core/shell tags with these nanogaps, also known as gap-enhanced Raman tags, have consequently garnered considerable attention due to their stability and lesser susceptibility to environmental conditions, thereby assuring a consistent SERS signal intensity. 42However, the photothermal effects present in plasmonic metal NPs cannot be overlooked, as they have the potential to disrupt or even obliterate biological samples during the imaging process.To mitigate this issue, Ye and coworkers 43 engineered an ultra-bright SERS probe.This probe was composed of noble metal Au@Ag core-shell rod-like nanocrystals (RNM) encapsulating Raman reporter molecules.The researchers prepared SERS probes with non-resonant effects by regulating the thickness of the Ag shell, thereby alleviating the detrimental photothermal effects on biological samples.

BIOMEDICAL APPLICATIONS OF SERS DETECTION
SERS is a non-invasive, rapid, and highly sensitive detection method. 47,48The technique can detect characteristic fingerprint peaks for different types of cells, proteins, 49 miRNA, 50 and DNA. 51Therefore, SERS has unique advantages in the detection of biological samples. 52SERS offers two methodologies regarding biomedical applications, including labeled detection with Raman reporter and recognition units and label-free detection without any tags. 53Related studies have found that both labeled and label-free SERS detection can detect circulating tumor cells (CTCs), exosomes, and drugs with high sensitivity.However, in vivo and in vitro SERS bioimaging is currently only possible with SERS probes with reporter tags.Compared with conventional methods such as ELISA, Western blot, immunohistochemistry (IHC), fluorescence detection, and imaging examination, SERS has the advantages of high sensitivity (up to 10 −15 M), fast detection speed, no photobleaching, less sample requirement, and simultaneous detection of multiple samples.In addition, with the advent of hand-held Raman devices, immediate intraoperative or bedside SERS detection is possible, which greatly increases the prospects for clinical translation of this technology.However, it should also be noted that SERS has the disadvantages of low penetration depth and limited field of view in biological imaging.

SERS detection for CTCs
CTCs are present in the blood and can be an essential reference for early tumor diagnosis.However, CTCs are usually extremely rare in peripheral blood (1 mL of blood contains a few CTCs).In addition, the presence of large numbers of leukocytes (10 7 ) and erythrocytes (∼5 × 10 9 ) can interfere with CTC measurements.Therefore, the development of sensitive CTC assays is essential.Wu et al. reported composite NPs with SERS functionality constructed from GNPs of different shapes (spheres, stars, rods) that can be used for the sensitive detection of CTCs. 54In their method, 4-mercaptobenzoic acid was utilized as a reporter for these differently shaped GNPs, endowing them with robust SERS signals.These NPs were subsequently modified with reduced bovine serum albumin to minimize nonspecific capture or uptake by healthy cells present in the blood.Finally, the NPs were deployed with folate (FA) ligands intended for targeting a variety of CTCs, which express folate receptor α (FRα) at high levels.This method is applicable for detecting tumor cells in blood samples, with a limit of detection (LOD) as low as 3 cells/mL.To further optimize the detection capability of SERS in complex environments, the group also developed a CTC analysis system based on triangular silver nanoprisms and superparamagnetic iron oxide NPs. 55The system can conduct CTC capture, enrichment, detection, and release functions (Figure 3A).Cancer cells such as ovarian, kidney, breast, and lung cancers with high FA expression can be targeted by constructing FR-targeting SERS NPs.The method's detection limit can be as low as 1 cell/mL.In addition, by adding excess free FRs, the enriched CTCs can also be released for further cell expansion, phenotype identification, and molecular analysis.

SERS detection of exosomes
Exosomes are small vesicles secreted by cells that appear in body fluids, such as blood, urine, and saliva.Compared to traditional invasive histopathological biopsies, exosomes offer the advantage of being cost-effective and minimally invasive in disease diagnosis, treatment, and monitoring. 56,57To effectively detect exosomes, the researchers have approached this in two ways. 49One is to develop excellent SERS substrates to enhance the SNR of single-molecule Raman spectroscopy.The other is to develop SERS labels modified with specific recognition elements to enhance the SERS response of the target.Sivashanmugan et al. reported bimetallic nanoplasma gapmode SERS substrates to study exosomes.Gold nanorods and silver nanocubes formed a strong plasma cavity effect. 58The optimized bimetallic nanoplasmonic gapmode SERS substrate, in conjunction with the SERS label-free detection method, was employed for the identifi-cation and detection of lung cancer cells.Remarkably, this approach allows the detection of tumor exosomes at concentrations 10 4 -10 5 times lower than those typically found in standard blood samples.To evaluate patient response to anti-programmed Death Ligand-1 (PD-L1)/programmed Death Receptor-(PD-1) immunotherapy, Xiao and coworkers developed a simple and rapid procedure for quantifying exosomal PD-L1 biomarkers from clinical serum samples (Figure 3B). 59Using Fe 3 O 4 @TiO 2 NPs, exosomes could be enriched and isolated from the solution in less than 5 min with a capture efficiency of 96.5%.After anti-PD-L1 antibody-modified SERS labels are prepared, exosomal PD-L1 can be accurately quantified in 4 μL of undiluted serum with a detection limit as low as 1 PD-L1 exosome/μL.In addition, the assay process is significantly shorter and can be completed in less than 40 min.Analysis based on the SERS intensity score of each sample allows differentiation between patients with early and advanced non-small cell lung cancer as well as healthy populations.Fan and coworkers innovatively developed a sensitive, direct SERS aptasensor utilizing gold nanostar (AuNS)-modified molybdenum disulfide (MoS 2 ) nanocomposites. 60They assembled ROX-labeled aptamers (ROX-Apt) onto the MoS 2 -AuNS surface, serving as recognition probes with a specific binding affinity for the transmembrane protein CD63 on exosomes.Through the synergy of the Raman enhancement effect between AuNSs and MoS 2 nanosheets, a striking ROX Raman signal was obtained.The advanced SERS aptasensor, crafted by the team, could proficiently detect gastric cancer exosomes within the concentration range of 55-5.5 × 10 5 μL −1 and had a detection threshold of 17 μL −1 .Significantly, the SERS aptasensor demonstrated substantial stability, offering promising potential for clinical application.

SERS detection for drugs
In the field of life sciences and public safety, monitoring drug abuse is highly dependent on the quantitative analysis of various drug molecules.A reliable SERS method for the detection of mephedrone by preparing gold and silver NPs in combination with fractional factorial design showed that the relative standard deviation of the mephedrone-specific Raman peak was as low as 0.51%.The LOD was estimated to be approximately 1.6 μg/mL (9.06 × 10 −6 M), allowing rapid and simple in situ detection. 61Usually, the target sample is dispersed in a complex mixture with impurities, which poses a great challenge for the detection of drugs, especially for trace analytes. 62SERS technology integrates single molecule sensitivity and fingerprinting features of molecular vibrations with high resistance to interference and therefore holds great promise for detecting complex samples. 63In recent years, the SERS technique has become an excellent surface analysis tool for the detection of drugs in urine, saliva, or blood. 64,65For example, methamphetamine can be effectively detected using a microfluidic device that can diffuse the sample in saliva into a designated area, while significant SERS signal enhancement can be achieved by introducing a salt solution to promote the aggregation of silver NPs. 66A novel liquid interface SERS platform for detecting drugs in urine is illustrated (Figure 3C). 67yclohexane (CYH) is used as the extraction solvent to extract drugs from urine with high extraction rates and less interference from impurities.CYH as the organic phase induces large-scale self-assembly of GNP arrays at the CYH/water interface.Self-directed arrays at the interface of two immiscible phases allow the detection of dispersed analytes such as methamphetamine and 3,4methylenedioxy-methamphetamine in both oil and watersoluble phases.The application of SERS detection has been greatly developed through the multiplex detection of such ultra-trace analytes and the two-phase dual analyte detection.

SERS imaging for in vitro experiment
Confocal Raman microscopy enables high-resolution cellular imaging based on biomolecules and has been widely used to obtain 3D information.Due to the extremely low SERS signal of endogenous biomolecules, monitoring and imaging are usually performed with the help of Raman signaling molecules.These Raman signaling molecules can be adsorbed onto small-sized SERS substrates, engulfed by cells, or immobilized on the surface.In one study, the authors prepared GNPs of approximately 42 nm, modified with a near-infrared laser-sensitive Raman dye and a cell-penetrating peptide. 68Such a system not only ensures that cells are successfully captured but also successfully images cells without photobleaching effects, as shown in Figure 4A. Figure 4B shows the imaging results for different cellular components of targeted cells at other incubation times, demonstrating the promising applications of SERS imaging encoded by Raman signaling molecules.
To increase the spectral stability of cellular imaging, Wu developed a semiconducting black TiO 2 NP (B-TiO 2 ) with significant SERS activity. 69By constructing a B-TiO 2 bioprobe targeting cancer cells to collect SERS profiles, rapid imaging of breast cancer cells can be achieved to meet the requirements for early screening and diagnosis of cancer cells.

SERS imaging for in vivo experiment
During tumor removal, imaging of the removed tissue is required to identify residual tumors at the margins and guide its complete removal.The feasibility of using SERS probes to detect single biomarkers after local staining of isolated mouse tissue with targeted gold-silica NPs for 1 h has been demonstrated. 70To further explore multiplex imaging of isolated tissues by SERS, a variety of antibody-modified NPs were stained against various tumor xenografts.As shown in Figure 4C, the multiple tumors were clearly distinguishable by imaging and EGFR and HER-2 expression showed excellent quantitative agreement with the corresponding flow cytometry results (R > 0.98). 71In addition to cell imaging and tissue imaging, the application of in vivo imaging is also an important exploration of SERS technology.To assess the multiplexed in vivo imaging capabilities of the constructed SERS probes, five SERS nanotags with less spectral overlap (S420, S421, S440, S466, and S470) were injected into mice via the tail vein.As the particles are taken up by Kupffer cells of the reticuloendothelial system, accumulation in the liver can be observed and deep tissue multiplexing can be assessed.As can be seen in Figure 4D, different SERS markers yield different imaging brightness, which informs the choice of in vivo imaging. 72The imaging technique correlates with the spectral stability of the designed SERS tag, the Raman signal enhancement capability, and biocompatibility.Exploring the construction of stable and excellent SERS probes will be an important prerequisite for broadening imaging applications.Di et al. 73 successfully fabricated a cisplatin-loaded gap-enhanced Raman marker using gold, mesoporous silica, a 1,4-benzenedithiol (1,4-BDT) Raman reporter, and cisplatin.This marker was employed for intraoperative detection and photothermal therapy (PTT) of advanced, unresectable, and disseminated ovarian tumors, notably enhancing the detection rate of minute disseminated tumors and improving prognosis.Similarly, Xiao and coworkers 74 prepared a gapenhanced Raman probe with a core-shell structure using plasma gold, a mesoporous silica layer, and the Raman reporter BDT.This probe was effectively used for the highly sensitive and accurate detection of prostate cancer during and after surgery.Moreover, its impressive photothermal performance, when combined with a highenergy laser, can be harnessed as an effective treatment for residual or recurrent tumors after the surgery.However, due to strong optical scattering and absorption in biological tissues, the use of Raman spectra for in vivo biomedical detection of deep lesions is typically constrained by the depth of tissue penetration.Spatially offset Raman spectroscopy (SORS) devices surmount these hurdles by maintaining a specific distance between the light source and the detector, thereby harnessing the high scattering nature of biological tissues to capture photons from deeper tissue layers. 75Consequently, the SORS configuration diminishes the background signal from the superficial layer, thereby amplifying the SNR of the signal from the deeper layer.There have been reports of successful noninvasive in vivo detection of glioblastoma tumors through mouse skulls via SORS technology.Additionally, transmission Raman spectroscopy (TRS) is a distinct form of the spatially offset Raman setup, wherein the laser and the Raman detector are positioned on opposing sides of the sample.TRS devices have reportedly accomplished noninvasive detection, demonstrating high tissue penetration and a probing depth extending up to 14 cm in ex vivo tissue experiments. 76,77Despite these clear advantages of TRS, its detection depth remains less than the thickness of the human body, the underlying mechanism is unclear, and the requirement for high-intensity laser excitation raises biological safety concerns, making the clinical translation of TRS a challenging endeavor. 78o enhance the targeted recognition capability of SERS biological probes and minimize accumulation at nontarget sites, generally, three strategies are employed: first, monoclonal antibodies are loaded onto SERS biological probes to bind with ligands or receptors on tumor cell membranes, thereby converting passive into active targeting.Second, surface modifications are applied to SERS bioprobes to diminish early phagocytosis by the endothelial reticular system in vivo, consequently increasing the number of nanomaterials reaching the target cells.Third, the physicochemical properties of the SERS biological probes are improved, including particle size reduction and potential alteration, thus indirectly augmenting the phagocytic capacity of target cells toward the material.

4.5
Integrated SERS detection platform

Lateral flow platforms for SERS detection
The application of SERS sensing technology in lateral flow test strips has also received attention, which also called paper-based microfluidics.A typical test strip contains four main structures, namely a sample pad, a conjugate pad, a nitrocellulose membrane, and an absorbent pad.There are two lines on the nitrocellulose membrane, referred to as the test and control lines.During the test, a liquid sample solution is dropped onto the sample pad, which is carried into the absorbent pad.During this process, the sample will bind to the conjugate (usually an antibody) in the conjugate pad.Finally, the bound target complexes will be captured at the test line and the others will reach the control line. 79In a study on the detection of disease-related markers, Zhao and coworkers proposed a lateral flow assay (LFA) based on core-shell SERS nanotags for the multiplexed and quantitative detection of cardiac biomarkers for the early diagnosis of acute myocardial infarction (Figure 5A). 80Nile blue A was embedded as a reporter molecule in silver-core and gold-shell NPs that collectively form the SERS label (Ag NBA @Au).Three test lines were used in the strips to detect three cardiac biomarkers, Myo, cTnI, and CK-MB.The authors noted that the amplified signal of the SERS nanotags (antibody-conjugated Ag NBA @Au) and the high surface area-to-volume ratio of the porous nitrocellulose membrane enabled ultrasensitive quantification of protein markers down to as low as 0.55 pg/mL of protein markers.This combined SERS-LFA is ultra-sensitive, fast, cost effective, and easy to use, while eliminating the need for sample pre-treatment and specialist technicians.

Microfluidic platforms for SERS detection
Microfluidics, also known as lab-on-a-chip, is a technology that integrates macro-reactors on tiny devices, having the advantages of low sample consumption, short reaction time, high detection efficiency, and high portability.Usually, the volume of biological reactions is small, which has facilitated the development of microfluidics in the field of bioanalysis.For SERS assays, the combination with the microfluidic platform can significantly improve detection efficiency and enable point-of-care.In a diagnostic study (Figure 5B), the authors developed a method to isolate plasma exosomes using microfluidics and analyze various exosome biomarkers using SERS technology to diagnose osteosarcoma. 81The microfluidic device based on tangential flow filtration allows for efficient separation of exosomes from cell supernatants and plasma samples.Properly designed SERS probes based on protein biomarkers achieve detection limits as low as two exosomes per microliter.Combined with multivariate statistical analysis, this enables differentiation between osteosarcoma patients and healthy controls with an accuracy of 95%.Combined with microfluidic SERS assays that do not require complex manipulation or expensive equipment, they hold great promise as a liquid biopsy technique for the clinical diagnosis of cancer.In addition, microfluidic SERS technology can be used for the detection and differentiation of a wide range of cancers, such as breast cancer, cervical cancer, and prostate cancer, and has excellent potential in the field of early cancer diagnosis. 82,83More importantly, microfluidic SERS technology also plays a role in point-of-care diseases such as blood glucose monitoring. 84

SERS FOR PRECISION DIAGNOSIS OF PANCREATIC CANCER
In recent years, SERS detection results of several studies have demonstrated the great potential of SERS technology in the early diagnosis and monitoring of the therapeutic efficacy of PC, 85,86 as shown in Table 1.It is expected to become a new detection tool in clinical practice.

Labeled SERS for the detection of PC
Due to the inherent stability and distinctiveness of the Raman spectrum peak exhibited by the reporter molecule, the labeled SERS method commonly employs Raman reporter molecules as SERS tags for achieving ultrasensitive detection of biomedical samples. 87Through TA B L E 1 Surface-enhanced Raman spectroscopy (SERS) bioprobes for early detection and monitoring of the therapeutic effect of pancreatic cancer (PC).SERS-Tag and specific antibodies binding, tumor cells can be targeted for recognition and capture. 88Simultaneously, using noble metal substrates can amplify the Raman signal emitted by captured tumor cells, thereby enabling highly sensitive diagnosis of these cells. 89Wang et al. developed a SERS-based sandwich immunoassay for the early diagnosis of PC (Figure 6). 90A SERS bioprobe was prepared using GNPs, 4-nitrophenyl thiol, monoclonal antibodies, and combined gold capture substrate to detect the tumor marker Mucin-4 (MUC4), which was impossible to detect in PC patient sera by ELISA.The research has also found that this SERS-based sandwich immunoassay is more effective than traditional methods in detecting the classic tumor marker carbohydrate antigen 19-9 (CA19-9) in 10 patients' serum (tumor stage IV).

Component
Similarly, Kim and coworkers reported a novel multiplex immunoassay that uses GNPs, 4-nitrobenzenethiol (4-NBT) and antibodies as a SERS bioprobe to detect the biomarkers CA19-9, matrix metalloproteinase-7 (MMP7), and MUC4 in serum samples. 91The innovative approach enables PC diagnosis by comparing the expression patterns of these biomarkers in five PC patients, pancreatitis patients (tumor stage IV), and 10 healthy individuals.Additionally, Porter and coworkers prepared a multiplexed assay platform using GNPs and 5,5-dithiobis(succinimidy1-2-nitrobenzoate) (Figure 7). 92The result of using this platform for the detection of two serum markers specific for PC, serum CA19-9 and MMP7, has the following advantages over conventional ELISA: (1) it is suitable for analytical multiplexing, ( 2) it requires only one-tenth of the sample volume required for ELISA, and ( 3) it has a superior LOD than ELISA.To investigate the ability of SERS to detect precancerous cells.Kircher and coworkers have developed a highly sensitive SERS bioprobe by coating AuNSs with silica and resonant Raman reporter IR-780 perchlorate. 93Using a mouse model of the pancreas in situ, they have effectively showcased the capability of AuNSs SERS bioprobe and its mapping modality in precisely identifying the presence and location of pancreatic tumors, including sub-millimeter-sized tumor metastatic lesions (Figure 8).The findings of this study also highlight the precise delineation of tumor tissue boundaries achievable through the utilization of the SERS MAPPING technique.In recent developments, SERS bioprobes have been employed for the detection of miRNA biomarkers and tumor-derived exosomes.Li et al. designed and prepared a 3D array of ordered micro-nanostructures based on AgNPs as a highly sensitive SERS substrate, followed by the addition of a rapid enrichment strategy magnetic beads @ exosome @ SERS detection probe (MEDP) to form an efficient sandwich immune complex for the detection of exosomes Leucine-rich-alpha-2-glycoprotein-1 Exosomes (LRG1) and Glypican-1 Exosomes (GPC1) for the diagnosis of PC (Figure 9). 86The MEDP@H-SERS immunocomplex presented a robust and dependable method for early PC diagnosis, exhibiting impressive sensitivity, specificity, and area under the curve values of 91.4%, 86.7%, and 0.95 (95% confidence interval: 0.849-0.991),respectively.Pang et al. 94 have fabricated a dual SERS bioprobe using Fe 3 O 4 @Ag-DNA and Au@Ag@5,5'-dithiobis- (2-nitrobenzoic acid)(DTNB) (Figure 10).The research found that this dual SERS-enhanced bioprobe could recognize low-level single base mismatches, even at concentrations as low as 1 aM.Furthermore, this experiment showed that using this SERS bioprobe to determine microRNA-10b concentrations in patient blood samples and residual plasma supernatants makes it possible to accurately distinguish between PC (tumor stage 1/II), chronic pancreatitis and healthy people.Li et al. reported a sandwich structure using a "chip exosome-PEARL tag" nanostructure (Figure 11). 95The nanostructure was formed by PDA modification of antibody reporter Ag (shell)-Au (core) multilayer (PEARL) SERS tags.This method mainly uses SERS immunosensors to target and identify migration invasion factor, a tumor biomarker highly expressed in the exosome of PC, for diagnosis.This method requires only 2 μL blood samples to diagnose and classify PC.This study provides a new strategy for the early diagnosis and staging of PC.

Label-free SERS bioprobe for the detection of PC
Label-free SERS detection effectively differentiates various types and subtypes of pancreatic tumor cells by analyzing the inherent spectral features of live cells. 96However, it is currently challenging to find a simple method to extract the characteristic Raman fingerprint peaks of PC cells, thus limiting the medical application of this technique.Many attempts have been made to overcome this limitation are summarized below: Ito et al. explored a highly sensitive method for diagnosing pancreatic tumors (tumor stage II-III). 97They placed five human serum samples on phosphor bronze chips with silver nano-hexagonal columns and irradiated them with a low-intensity helium-neon red laser beam at 633 nm.Tumor cells were successfully identified by analyzing the SERS spectra and peak intensities of the samples.Remarkably, this method demonstrated the ability to detect tumor cells at concentrations as low as 1/1000.Subsequent experiments also confirmed that this method is effective for detecting gastrointestinal tumors.Kaur and coworkers 85 reported label-free detection and analysis of exosomes from normal and PC cells using SERS combined with principal component difference function analysis (PC-DFA), with Ti/Au as the substrate (Figure 12).This method utilizes label-free SERS detection in conjunction with PC-DFA to establish a predictive model.Through this model, the accuracy for identifying PC patients and healthy individuals is 87% and 90%, respectively.Lee and coworkers 98 developed a reliable and rapid SERS-based urine analysis platform using 3D-stacked silver nanowires on a glass fiber filter sensor (Figure 11).The platform performed label-free urine-SERS analysis to obtain spectral data, followed by unsupervised principal component analysis (PCA) and supervised orthogonal partial least squares discriminant analysis (OPLS-DA) for multivariate analysis to differentiate pancreatic tumor patients from healthy individuals.Furthermore, this study found that the detection of PC using the OPLS-DA method was superior to PCA, with sensitivity and specificity reaching 100%.

SERS bioprobe for monitoring the therapeutic effect
Accurate monitoring of anti-tumor drug concentrations is crucial for effective tumor therapy.Excessive drug concentrations can lead to unnecessary harm to normal cells and severe side effects, while insufficient drug concentrations may fail to eliminate tumors effectively.In recent years, several studies have found that assessing the concentration and efficacy of anti-tumor drugs through changes in SERS signal intensity is feasible and accurate.Litti et al. 99 synthesized a new SERS-based colloidal nanosensor for monitoring the concentration of erlotinib, a drug targeted to therapy for pancreatic or lung cancer.They used erlotinib and propynyl fluorescent red (PFR), which competed with AuNP-N3, to produce a SERS signal to demonstrate changes in PFR's SERS corresponding to changes in erlotinib concentration.The results of this study provide a new method and idea for detecting anti-cancer drug concentrations.Subsequently, Mukhopadhyay and coworkers 100 successfully developed a highly specific and sensitive DNA-Single-walled carbon nanotube (SWCNT) SERS bioprobe for real-time monitoring of the response of PC cells to chemotherapy drugs such as gemcitabine.The reversible alteration of the Raman G band in the carbon nanotubes within the SERS bioprobe can reflect the concentration changes of hydrogen peroxide within tumor cells.As gemcitabine functions by generating a large amount of hydrogen peroxide to kill PC cells, the real-time changes in hydrogen peroxide concentration within PC cells can indirectly reflect the response of PC to gemcitabine.Furthermore, Sujai et al. 101 reported a photothermal therapeutic nanoenvelope (MnO 2 @AuNPs).This nanoenvelope, composed of manganese dioxide-covered AuNPs, exhibited excellent photothermal therapeutic efficacy when exposed to 808 nm laser irradiation for killing PC cells (Figure 13).Notably, the therapeutic effect of PTT could be monitored

CONCLUSION AND FUTURE PERSPECTIVE
In this review, we aim to provide a comprehensive overview of the recent advancements in biomedical applications of SERS.SERS spectroscopy is widely used in the biological field due to its advantages of high sensitivity, rapid analysis, fingerprint spectroscopy, and noninvasiveness.
In tumor precise diagnosis, SERS bioprobes can accurately diagnose individual tumor cells through labeled and label-free detection.In tumor precision therapy, the initial therapeutic effect of anti-tumor drugs can be monitored by observing the changes in the intensity and peak position of SERS fingerprint peaks.The synthesis method for PC-specific labeled SERS bioprobes is usually divided into two steps.Initially, monoclonal antibodies are attached to NPs that carry reporter molecules and SERS substrates.This approach facilitates the precise targeting and capture of specific markers, forming a sandwich structure.Subsequently, the SERS signal is obtained using laser irradiation through reporter molecules on the NPs, thus indirectly detecting PC cells in body fluids such as blood and urine.Tumor cells often secrete or express a variety of different types of biomarkers.As a result, the detection sensitivity of a SERS biosensor increases with the ability to simultaneously identify a greater number of tumor biomarkers.Therefore, the development of a PC SERS bioprobe that combines multiple antibodies to simultaneously identify multiple biomarkers is one of the important future research directions.Kim and coworkers 91 have performed a preliminary study in this area using GNPs, 4-NBT, and antibodies to prepare a multiplex immunoassay with an LOD of 2 ng/mL as a SERS bioprobe to accurately detect three PC biomarkers CA19-9, MMP7, and MUC4 simultaneously.
In comparison, label-free SERS detection is a direct analysis of the SERS signals of different cells to identify tumor cells from normal cells based on the specific molecular fingerprint peaks of different cells. 102However, the SERS signals of individual cells are not very distinct and characteristic fingerprint peaks are not obvious, so it is often necessary to combine machine learning methods to improve the sensitivity and accuracy of the test. 1035][106] In addition, the composition and structure of the SERS substrate have a more significant impact on the effectiveness of the detection in label-free SERS detection than in labeled SERS detection.Currently, commonly used substrates are solidphase and sol-gels composed of gold or silver NPs. Lee and coworkers 98 prepared a SERS substrate for detecting PC cells using 3D-stacked silver nanowires and GFFs.The substrate's EF was up to 1.7 × 10 7 .Besides the material's structure, the substrate material's composition also significantly impacts the performance of the SERS detection.Currently, SERS substrates are mainly composed of gold or silver, while SERS substrates made of non-noble metal nanomaterials (e.g., C, Ti, Zn, Cu, Mo, and W) offer better economy, stability, selectivity, and biocompatibility than precious metal substrates.In addition, recent studies have found that semiconductor materials have good SERS performance.Li et al. 107 prepared SERS substrates of SnSe 2 NPAs, which demonstrated ultra-low detection limits (1 × 10 −12 M), high EFs (1.33 × 10 6 ), and excellent homogeneity (relative standard deviation down to 7.7%), meeting or exceeding the performance of conventional metal SERS substrates.It is one of the most sensitive semiconductor SERS substrates reported to date.However, the use of nonprecious metal and semiconductor SERS substrates for detecting PC has not been reported.
Microfluidic platform-based exosome SERS detection of PC is also one of the future research hotspots. 108Exosomes can directly reflect the fundamental information of PC cells, making them potential biomarkers for early diagnosis of PC. 109 However, due to the small size of exosomes and their relative density to body fluids, it is difficult to isolate and analyze exosomes from complex biological samples. 110Traditional exosome isolation methods require large instruments and equipment, which are time consuming and complicated.Microfluidics has unique advantages in terms of exosome isolation, enrichment and SERS detection, so combining the three techniques can further improve the sensitivity and accuracy of early diagnosis of PC. 111 In addition, the microfluidic technology can be used for SERS detection and analysis of single PC cells, laying a solid foundation for SERS detection in the pancreatic tumor microenvironment, tumor metastasis, and tumor immunity.
The detection of pancreatic tumor metabolites using SERS is a burgeoning area of research.SERS probes that are responsive to tumor metabolites, such as those that respond to pH changes, reactive oxygen species, and enzyme activity, can aid in the precise identification of tumor metastasis and invasion in cancers such as hepatic and glioma. 112Moreover, study reported by Lu and coworkers combined creatine, inosine, beta-sitosterol, sphinganine, and glycocholic acid into a panel biomarker, which may improve the trace-based diagnosis of PC in clinical samples.This diagnostic approach significantly outperformed traditional markers including Carbohydrate antigen 125 (CA125), CA19-9, CA242, and Carcinoembryonic antigen (CEA). 113Another research published by Yin and coworkers developed a non-invasive method for detecting PC by employing machine learning-assisted metabolomics.The team used a support vector machinegreedy algorithm and high-resolution mass spectrometry to analyze untargeted metabolomics data and identified 17 serum metabolic markers.They then established a targeted metabolic detection method based on a multiple reaction detection mode, as well as an artificial intelligence disease classification model using liquid chromatography-mass spectrometry.This method exhibits over 85% accuracy, thus significantly surpassing CA19-9 and computed tomography examination in terms of detection efficiency. 114urthermore, Qian and coworkers developed a coreshell structured Fe 3 O 4 @SiO 2 @Pt particle nanoreactor for the analysis of tumor metabolites.By using the optimal composition of this nanomaterial, they achieved 84% sensitivity and 92% specificity, thereby demonstrating its effectiveness as a rapid diagnostic tool for PC. 115Consequently, it is plausible to utilize metabolite-responsive SERS probes, with or without tags, for detecting pancreatic tumors.
Real-time surface-enhanced Raman-guided surgical resection of tumors is an important future medical application of Raman technology.The technique distinguishes between tumor and normal tissue by analyzing in real time the specific Raman signal produced by the tumor cells themselves or by secretions, thus delineating the tumor boundary precisely for precise tumor removal. 116The technology has been used to conduct initial exploratory research in surgical navigation strategies for glioma Li and coworkers used AuNSs to prepare a surface-enhanced resonance Raman scattering bioprobe with a PH ratiometrically response across the blood-brain barrier and proposed a novel strategy to observe acidic "metabolic boundaries" to guide the surgery.The strategy was subsequently demonstrated in an animal glioma precision resection experiment. 117In recent years, a series of studies have demonstrated that it is feasible and safe to use the endoscopic Raman systems for accurate early qualitative diagnosis and real-time tumor margin identification of gastric, 118 esophageal, 119 and colorectal cancers. 120ither surgical navigation or endoscopic systems play an extremely important role in the diagnosis and treatment of PC.Hence, it is reasonable to assume that combining SERS technology with intraoperative navigation systems or SERS technology with endoscopic technology to improve the early diagnosis and survival rates of PC is feasible shortly and is one of the research directions that deserves our focused attention.

F I G U R E 1
Electromagnetic and chemical mechanism for surface-enhanced Raman spectroscopy (SERS) enhancement.A, Electromagnetic enhancement in SERS based on plasmonic nanospheres.B, Schematic illustration of a "hot spot" in the gap between adjacent particles and the corresponding change in SERS enhancement factor with relative positions.Comparison of the charge transfer transitions in a metal-molecule system C, and a semiconductor-molecule system D. Adapted and reproduced from ref. 20 with permission from the Elsevier Copyright 2020.

F I G U R E 2
Examples of noble metal and semiconductor surface-enhanced Raman spectroscopy (SERS) substrate fabrication: A, the synthesis process of Ag/anodic aluminum oxide (AAO)-2 substrate; adapted and reproduced from ref. 37 with permission from the Royal Society of Chemistry (RCS) Copyright 2011.B, The fabrication process of Au or Au/Ag nanodisk array; adapted and reproduced from ref. 38 with permission from the IOP Publishing Copyright 2011.C, Schematic illustration of the formation of Cu crystal; adapted and reproduced from ref. 44 with permission from the John Wiley and Sons Copyright 2018.D, Self-assembly process for the formation of Cu 2 O cube-like superstructures; adapted and reproduced from ref. 45 with permission from the John Wiley and Sons Copyright 2016.

F I G U R E 3
Example of surface-enhanced Raman spectroscopy (SERS) detection: A, schematic illustration of the designed silver nanoprisms (AgNPR) and superparamagnetic iron oxide nanoparticles (SPION) to form a supersensitive circulating tumor cell (CTC) detection SERS bioprobe, with the functions of capture, enrichment, detection, and release of CTCs in blood; adapted and reproduced from ref. 55 with permission from the American Chemical Society (ACS) Copyright 2018.B, Schematic view of the nanoparticles synthesis and SERS tag-based exosomal PD-L1 detection; adapted and reproduced from ref. 59 with permission from the Elsevier Copyright 2019.C, Schematic illustrations and optical images of the urine extract-induced self-assembly of gold nanoparticle (GNP) arrays at the liquid/air interface for SERS detection and microscope images of self-assembly process of the large-scale interfacial GNP arrays; adapted and reproduced from ref. 65 with permission from the American Chemical Society (ACS) Copyright 2016.

F I G U R E 4
Examples of surface-enhanced Raman spectroscopy (SERS) imaging.A, Synthetic scheme of Raman dye (44DP)-coded gold nanoparticles with uniform intra-nanogap (Au-NNPs).The solution color and high-resolution transmission electron microscopy (HR-TEM) image of 44DP-coded Au-NNPs prepared from A10 spacer DNA-AuNP, G10 spacer DNA-AuNP, C10 spacer DNA-AuNP, and T10 spacer DNA-AuNP.Raman spectra of 44DP-coded Au-NNP solution prepared from four different spacer DNA with an excitation of 633 and 785 nm.B, Time-dependent live cell Raman images after incubation with subcellular targeting NNPs; adapted from ref. 68 with permission from American Chemical Society (ACS) Copyright 2015.C, Quantitative molecular phenotyping (QMP) imaging, with 0.5-mm spatial resolution, of tumor-xenograft specimens stained with a three-flavor nanoparticle (NP) mixture (EGFR-NPs, HER2-NPs, and isotype-NPs); adapted from ref. 71 with permission from Springer Nature Copyright 2016.D, Graph depicting five unique Raman spectra, each associated with its own SERS batch: S420 (red), S421 (green), S440 (blue), S466 (yellow), and S470 (orange).Deep-tissue detection Raman image of liver overlaid on digital photo of mouse in the liver after 24 h post i.v.injection; adapted from ref. 72 with permission from the National Academy of Sciences of the United States of America Copyright 2009.

F I G U R E 5
Example of integrated surface-enhanced Raman spectroscopy (SERS) detection platform.A, Schematic illustration of the core-shell SERS nanotag-based multiplex lateral flow assay (LFA); adapted from ref. 80 with permission from the Elsevier Copyright 2018.B, Schematic illustration of SERS profiling of three biomarkers on plasma-derived exosomes isolated by size-dependent microfluidic filtration for the diagnosis of osteosarcoma; adapted from ref. 81 with permission from the Elsevier Copyright 2022.
Photothermal therapy

F I G U R E 7 A
, Schematic of surface-enhanced Raman spectroscopy (SERS) array for matrix metalloproteinase-7 (MMP7) and carbohydrate antigen 19-9 (CA19-9).B, Enzyme-linked immunosorbent assay (ELISA) calibration plots for the same MMP7 and CA19-9 standards used in A,.C, SERS calibration plots for MMP7 and CA19-9 antigen spiked into 1:4 pooled human serum:phosphate-buffered saline (PBS) buffer.Adapted and reproduced from ref. 91 with permission from the Royal Society of Chemistry.F I G U R E 8 A, Schematic and three-dimensional representations of the surface-enhanced Raman spectroscopy (SERS) nanostar geometry.Transmission electron micrographs of a single SERS nanostar and a population of SERS nanostars are shown.B, In situ photograph and corresponding Raman images of the exposed upper abdomen in a mouse with a Pancreatic Ductal Adenocarcinoma (PDAC) in the head of the pancreas and other normal-appearing regions.C, Hematoxylin and eosin (H&E) staining of the whole pancreas, including PDAC (arrow 1) and PanIN (arrow 2).Histology and keratin 19 (KRT19) staining in regions 1 and 2 confirmed lesions.Insets are 4× magnification views.Adapted and reproduced from ref. 93 with permission from the American Association for the Advancement of Science Copyright 2015.

F
I G U R E 9 A, The synthesis process of the Hierarchical SERS (H-SERS) substrate.B, Receiver operating characteristic (ROC) analysis to evaluate the diagnostic power of the Glypican-1 Exosomes (GPC1-Exos) and Leucine-rich-alpha-2-glycoprotein-1 Exosomes (LRG1-Exos) for the detection of early PaC.C, Illustration of the construction of exosome capture system, the fabrication of surface-enhanced Raman spectroscopy (SERS) detection probes, and the SERS detection of exosomes.Adapted and reproduced from ref. 6, 86 with permission from the John Wiley and Sons Copyright 2022.F I G U R E 1 0 A, The synthesis process of Fe 3 O 4 @Ag-SERS tags.B, Determination of microRNA-10b concentration in exosome and residual supernatant plasma from three patients with PDAC, three normal control (NC), and three patients with chronic pancreatitis (CP) using our surface-enhanced Raman spectroscopy (SERS) sensors.C, Scheme of SERS detection of microRNA.Adapted and reproduced from ref. 94 with permission from the Elsevier Copyright 2019.F I G U R E 1 1 A, The schematic of the PDA chip and PEARL surface-enhanced Raman spectroscopy (SERS) tag-based exosome sensors.B, Receiver operating characteristic (ROC) curves were calculated for single exosome markers (migration invasion factor [MIF], GPC1, and EGFR) (red: pancreatic cancer vs. healthy controls; purple: metastasis vs. non-metastasis; and green: P1, P2 vs. P3).AUC stands for the area under the curve.Adapted and reproduced from ref. 95 with permission from the Royal Society of Chemistry Copyright 2018.

F
I G U R E 1 2 A, Schematic diagram of the silver nanowire (AgNW)-glass fiber filter (GFF) surface-enhanced Raman spectroscopy (SERS) sensor and urine analysis for the diagnosis of pancreatic cancer.Principal component analysis (PCA) for B, three groups of normal control, pancreatic cancer, and prostate cancer, and for C, pancreatic cancer and prostate cancer.Orthogonal partial least squares discriminant analysis (OPLS-DA) for D, classification of normal control and cancer groups and for E, classification of pancreatic cancer and prostate cancer.Adapted and reproduced from ref. 98 with permission from the American Chemical Society (ACS) Copyright 2021.
in real-time by observing changes in cellular SERS signals.Evaluating tumor cell apoptosis and necrosis after PTT through changes in SERS signals within PC cells was consistent with the results obtained through flow cytometry.However, this assessment method based on changes in SERS signals requires less sample and time.It offers higher sensitivity, providing a new approach and F I G U R E 1 3 Schematic representation of photothermal therapeutic nanoenvelope (PTTNe) as a surface-enhanced Raman spectroscopy (SERS)-guided photothermal agent for pancreatic cancer.Adapted and reproduced from ref. 101 with permission from the Elsevier Copyright 2021.concept for accurate real-time detection of therapeutic efficacy.
46e unique 3D structures and resulting plasma coupling effect concentrated the "hot spots" at the tips and gaps of the crystals, resulting in strong SERS effects, as shown in Figure2C.The size distribution and alignment of the 3D sword-shaped Cu nanocrystals are relatively uniform, and the EF can reach 10 7 .Lin et al.45prepared a recrystallization-induced self-assembly strategy to fabricate a 3D cubic Cu 2 O superstructure SERS substrate enriched with Cu vacancies.The presence of Cu vacancy defects in this structure induces electrostatic adsorption and facilitates resonance coupling between the complexes, resulting in a synergistic enhancement of the charge transfer process and the SERS effect (Figure2D).The combined effects of Cu vacancy defects and resonance coupling significantly enhance the SERS effect.The EFs were measured at 8 × 8 for individual Cu 2 O superconstituent particles, while for individual Cu 2 O superstructure particles, the EFs reached an impressive value of 8 × 10 5 .Moreover, the detection limits for R6G and Crystal violet (CV) molecules were notably low, at 10 −9 M. Zhao and coworkers46synthesized MoO 3 -xH 2 O QDs to achieve direct and sensitive SERS fingerprinting of inorganic hydrazine hydrate.The synthesized QDs, with an average size of 2.2 nm, exhibited a significant improvement in detecting hydrazine hydrate compared to larger particles of 10 and 100 nm.The detection limit of the QDs was approximately 4 × 10 −5 M, reducing the minimum detectable concentration by at least 1000 times.