Artificial Intelligent Label-Free SERS Profiling of Serum Exosomes for Breast Cancer Diagnosis and Postoperative Assessment

Abstract

Breast cancer subtypes have important implications of treatment responses and clinical outcomes. Exosomes have been considered as promising biomarkers for liquid biopsies, but the utility of exosomes for accurate diagnosis of distinct breast cancer subtypes is a grand challenge due to the difficulty in uncovering the subtle compositional difference in complex clinical settings. Herein, we report an artificial intelligent surface-enhanced Raman spectroscopy (SERS) strategy for labelfree spectroscopic analysis of serum exosomes, allowing for accurate diagnosis of breast cancer and assessment of surgical outcomes. Our deep learning algorithm trained with SERS spectra of cancer cell-derived exosomes is demonstrated with a 100% prediction accuracy for human patients with different breast cancer subtypes who do not undergo surgery using SERS spectra of serum exosomes. Furthermore, when combined with similarity analysis by principal component analysis, our approach is able to evaluate the surgical outcomes of breast cancer of distinct molecular subtypes.

Integrated SERS-Vertical Flow Biosensor Enabling Multiplexed Quantitative Profiling of Serological Exosomal Proteins in Patients for Accurate Breast Cancer Subtyping

“…In fact, the four breast cancer cell lines have been widely utilized for modeling the intraheterogeneity of breast tumor at cellular levels. 48,55 Our results also suggest that the cancer cell-derived EXOs could be a good marker for breast cancer subtyping, consistent with our previous work. 55 In addition, the presented study unambiguously substantiates that the multiplexed quantitative analysis of a panel of exosomal proteins in human serums enables accurate breast cancer subtyping.…”

“…19,48 We isolated cancerous EXOs from their cell culture supernatant using a size-exclusion chromatography (SEC) technique. 55 Human serum EXOs were obtained with the commercial ExoQuick Exosome Isolation kit according to the manufacturer's instruction from blood serums of healthy individuals and human patients of luminal A/B, HER2+ and triple negative breast cancer subtypes. We performed comprehensive EXO analysis using TEM, nanoparticle tracking analysis (NTA), dynamic light scattering, and Western blot measurements.…”

“…When reaching 80% confluency, the cells were washed with phosphate buffered saline (PBS, pH 7.4) and transferred to serum-free DMEM medium. After 48 h, the culture medium was collected and utilized for isolation of cellular EXOs using a size-exclusion chromatography (SEC) technique . Specifically, the collected culture medium was centrifuged sequentially at 500 g for 10 min and 5000 g for 30 min to remove whole cells and dead cells.…”

Advanced technologies for molecular diagnosis of cancer: State of pre-clinical tumor-derived exosome liquid biopsies

“…However, this requires that such assays be able to detect a specific biomarker-associated Raman scattering signature within the spectra produced by the entire sample. Two groups have employed deep learning to generate model that can distinguish SERS signal produced by plasma/serum EVs from patients with early-stage lung adenocarcinomas (stage IA, IB and IIB) or breast cancer (triple negative, HER2+, luminal A and B) versus their corresponding non-malignant controls to provide proof-of-principle data for the potential utility of this approach [ 80 , 117 ]. However, another study employed a Raman dye to produce a consistent SERS signal for EVs that were specifically captured by specific antibodies for targeted EV biomarker proteins.…”

Dual-Aptamer-Assisted Ratiometric SERS Biosensor for Ultrasensitive and Precise Identification of Breast Cancer Exosomes

“…Multiple Raman probes or machine learning models are usually used in the SERS detections of exosomes to obtain a more accurate diagnosis, − but these methods tend to be cumbersome and their sensitivity is not satisfactory for the early diagnosis of cancers. Therefore, a more sensitive detection approach is urgently needed.…”