Background: Bladder cancer (BC) has the highest per-patient cost of all cancer types. Hence, we aim to develop a non-invasive, point-of-care tool for the diagnostic and molecular stratification of patients with BC based on combined microRNAs (miRNAs) and surface-enhanced Raman spectroscopy (SERS) profiling of urine. Methods: Next-generation sequencing of the whole miRNome and SERS profiling were performed on urine samples collected from 15 patients with BC and 16 control subjects (CTRLs). A retrospective cohort (BC = 66 and CTRL = 50) and RT-qPCR were used to confirm the selected differently expressed miRNAs. Diagnostic accuracy was assessed using machine learning algorithms (logistic regression, naïve Bayes, and random forest), which were trained to discriminate between BC and CTRL, using as input either miRNAs, SERS, or both. The molecular stratification of BC based on miRNA and SERS profiling was performed to discriminate between high-grade and low-grade tumors and between luminal and basal types. Results: Combining SERS data with three differentially expressed miRNAs (miR-34a-5p, miR-205-3p, miR-210-3p) yielded an Area Under the Curve (AUC) of 0.92 ± 0.06 in discriminating between BC and CTRL, an accuracy which was superior either to miRNAs (AUC = 0.84 ± 0.03) or SERS data (AUC = 0.84 ± 0.05) individually. When evaluating the classification accuracy for luminal and basal BC, the combination of miRNAs and SERS profiling averaged an AUC of 0.95 ± 0.03 across the three machine learning algorithms, again better than miRNA (AUC = 0.89 ± 0.04) or SERS (AUC = 0.92 ± 0.05) individually, although SERS alone performed better in terms of classification accuracy. Conclusion: miRNA profiling synergizes with SERS profiling for point-of-care diagnostic and molecular stratification of BC. By combining the two liquid biopsy methods, a clinically relevant tool that can aid BC patients is envisaged.

Background: We tested the feasibility of a simple method for assessment of prostate cancer (PCa) aggressiveness using diffusion-weighted magnetic resonance imaging (MRI) to calculate apparent diffusion coefficient (ADC) ratios between prostate cancer and healthy prostatic tissue. Methods: The requirement for institutional review board approval was waived. A set of 20 standardized core transperineal saturation biopsy specimens served as the reference standard for placement of regions of interest on ADC maps in tumorous and normal prostatic tissue of 22 men with PCa (median Gleason score: 7; range, 6–9). A total of 128 positive sectors were included for evaluation. Two diagnostic ratios were computed between tumor ADCs and normal sector ADCs: the ADC peripheral ratio (the ratio between tumor ADC and normal peripheral zone tissue, ADC-PR), and the ADC central ratio (the ratio between tumor ADC and normal central zone tissue, ADC-CR). The performance of the two ratios in detecting high-risk tumor foci (Gleason 8 and 9) was assessed using the area under the receiver operating characteristic curve (AUC). Results: Both ADC ratios presented significantly lower values in high-risk tumors (0.48 ± 0.13 for ADC-CR and 0.40 ± 0.09 for ADC-PR) compared with low-risk tumors (0.66 ± 0.17 for ADC-CR and 0.54 ± 0.09 for ADC-PR) (p < 0.001) and had better diagnostic performance (ADC-CR AUC = 0.77, sensitivity = 82.2%, specificity = 66.7% and ADC-PR AUC = 0.90, sensitivity = 93.7%, specificity = 80%) than stand-alone tumor ADCs (AUC of 0.75, sensitivity = 72.7%, specificity = 70.6%) for identifying high-risk lesions. Conclusions: The ADC ratio as an intrapatient-normalized diagnostic tool may be better in detecting high-grade lesions compared with analysis based on tumor ADCs alone, and may reduce the rate of biopsies.

Renal cancer (RC) represents 3% of all cancers, with a 2% annual increase in incidence worldwide, opening the discussion about the need for screening. However, no established screening tool currently exists for RC. To tackle this issue, we assessed surface-enhanced Raman scattering (SERS) profiling of serum as a liquid biopsy strategy to detect renal cell carcinoma (RCC), the most prevalent histologic subtype of RC. Thus, serum samples were collected from 23 patients with RCC and 27 controls (CTRL) presenting with a benign urological pathology such as lithiasis or benign prostatic hypertrophy. SERS profiling of deproteinized serum yielded SERS band spectra attributed mainly to purine metabolites, which exhibited higher intensities in the RCC group, and Raman bands of carotenoids, which exhibited lower intensities in the RCC group. Principal component analysis (PCA) of the SERS spectra showed a tendency for the unsupervised clustering of the two groups. Next, three machine learning algorithms (random forest, kNN, naïve Bayes) were implemented as supervised classification algorithms for achieving discrimination between the RCC and CTRL groups, yielding an AUC of 0.78 for random forest, 0.78 for kNN, and 0.76 for naïve Bayes (average AUC 0.77 ± 0.01). The present study highlights the potential of SERS liquid biopsy as a diagnostic and screening strategy for RCC. Further studies involving large cohorts and other urologic malignancies as controls are needed to validate the proposed SERS approach.

The lack of an accurate point-of-care detection system for microalbuminuria represents an important unmet medical need that contributes to the morbidity and mortality of patients with kidney diseases. In this proof-of-concept study, we used SERS spectroscopy to detect urinary albumin concentrations in the normal-to-mildly increased albuminuria range, a strategy that could be useful for the early diagnosis of renal impairment due to uncontrolled hypertension, cardiovascular disease or diabetes. We analyzed 27 urine samples by SERS, using iodide-modified silver nanoparticles and we could discriminate between groups with high and low albumin concentrations with an overall accuracy of 89%, 93% and 89%, using principal component analysis–linear discriminant analysis and cut-off values of 3, 6 and 10 μg mL⁻¹ for urinary albumin concentrations, respectively. We achieved a detection limit of 3 μg mL⁻¹ for human serum albumin based on the 1002 cm⁻¹ SERS band, attributed to the ring breathing vibration of phenylalanine. Our detection limit is similar to that of the immunoturbidimetric assays and around one order of magnitude below the detection limit of urinary dipsticks used to detect microalbuminuria. We used principal least squares regression for building a spectral model for quantifying albumin. Using an independent prediction set, the R² and root mean squared error of prediction between predicted and reference values of human serum albumin concentrations were 0.982 and 2.82, respectively. Here, we show that direct SERS spectroscopy has the sensitivity required for detecting clinically relevant concentrations of urinary albumin, a strategy that could be used in the future for the point-of-care screening of microalbuminuria.