Predictive model for early detection of type 2 diabetes using patients' clinical symptoms, demographic features, and knowledge of diabetes.

Background and Aims: With the global rise in type 2 diabetes, predictive modeling has become crucial for early detection, particularly in populations with low routine medical checkup profiles. This study aimed to develop a predictive model for type 2 diabetes using health check-up data focusing on clinical details, demographic features, biochemical markers, and diabetes knowledge. Methods: Data from 444 Nigerian patients were collected and analysed. We used 80% of this data set for training, and the remaining 20% for testing. Multivariable penalized logistic regression was employed to predict the disease onset, incorporating waist-hip ratio (WHR), triglycerides (TG), catalase, and atherogenic indices of plasma (AIP). Results: The predictive model demonstrated high accuracy, with an area under the curve of 99% (95% CI = 97%-100%) for the training set and 94% (95% CI = 89%-99%) for the test set. Notably, an increase in WHR (adjusted odds ratio [AOR] = 70.35; 95% CI = 10.04-493.1, p-value < 0.001) and elevated AIP (AOR = 4.55; 95% CI = 1.48-13.95, p-value = 0.008) levels were significantly associated with a higher risk of type 2 diabetes, while higher catalase levels (AOR = 0.33; 95% CI = 0.22-0.49, p < 0.001) correlated with a decreased risk. In contrast, TG levels (AOR = 1.04; 95% CI = 0.40-2.71, p-value = 0.94) were not associated with the disease. Conclusion: This study emphasizes the importance of using distinct clinical and biochemical markers for early type 2 diabetes detection in Nigeria, reflecting global trends in diabetes modeling, and highlighting the need for context-specific methods. The development of a web application based on these results aims to facilitate the early identification of individuals at risk, potentially reducing health complications, and improving diabetes management strategies in diverse settings.

Prediction of malaria positivity using patients' demographic and environmental features and clinical symptoms to complement parasitological confirmation before treatment.

BACKGROUND: Current malaria diagnosis methods that rely on microscopy and Histidine Rich Protein-2 (HRP2)-based rapid diagnostic tests (RDT) have drawbacks that necessitate the development of improved and complementary malaria diagnostic methods to overcome some or all these limitations. Consequently, the addition of automated detection and classification of malaria using laboratory methods can provide patients with more accurate and faster diagnosis. Therefore, this study used a machine-learning model to predict Plasmodium falciparum (Pf) antigen positivity (presence of malaria) based on sociodemographic behaviour, environment, and clinical features. METHOD: Data from 200 Nigerian patients were used to develop predictive models using nested cross-validation and sequential backward feature selection (SBFS), with 80% of the dataset randomly selected for training and optimisation and the remaining 20% for testing the models. Outcomes were classified as Pf-positive or Pf-negative, corresponding to the presence or absence of malaria, respectively. RESULTS: Among the three machine learning models examined, the penalised logistic regression model had the best area under the receiver operating characteristic curve for the training set (AUC = 84%; 95% confidence interval [CI]: 75-93%) and test set (AUC = 83%; 95% CI: 63-100%). Increased odds of malaria were associated with higher body weight (adjusted odds ratio (AOR) = 4.50, 95% CI: 2.27 to 8.01, p < 0.0001). Even though the association between the odds of having malaria and body temperature was not significant, patients with high body temperature had higher odds of testing positive for the Pf antigen than those who did not have high body temperature (AOR = 1.40, 95% CI: 0.99 to 1.91, p = 0.068). In addition, patients who had bushes in their surroundings (AOR = 2.60, 95% CI: 1.30 to 4.66, p = 0.006) or experienced fever (AOR = 2.10, 95% CI: 0.88 to 4.24, p = 0.099), headache (AOR = 2.07; 95% CI: 0.95 to 3.95, p = 0.068), muscle pain (AOR = 1.49; 95% CI: 0.66 to 3.39, p = 0.333), and vomiting (AOR = 2.32; 95% CI: 0.85 to 6.82, p = 0.097) were more likely to experience malaria. In contrast, decreased odds of malaria were associated with age (AOR = 0.62, 95% CI: 0.41 to 0.90, p = 0.012) and BMI (AOR = 0.47, 95% CI: 0.26 to 0.80, p = 0.006). CONCLUSION: Newly developed routinely collected baseline sociodemographic, environmental, and clinical features to predict Pf antigen positivity may be a valuable tool for clinical decision-making.

Classification of the tumor immune microenvironment and associations with outcomes in patients with metastatic melanoma treated with immunotherapies.

BACKGROUND: Tumor microenvironment (TME) characteristics are potential biomarkers of response to immune checkpoint inhibitors in metastatic melanoma. This study developed a method to perform unsupervised classification of TME of metastatic melanoma. METHODS: We used multiplex immunohistochemical and quantitative pathology-derived assessment of immune cell compositions of intratumoral and peritumoral regions of metastatic melanoma baseline biopsies to classify TME in relation to response to anti-programmed cell death protein 1 (PD-1) monotherapy or in combination with anti-cytotoxic T-cell lymphocyte-4 (ipilimumab (IPI)+PD-1). RESULTS: Spatial profiling of CD8+T cells, macrophages, and melanoma cells, as well as phenotypic PD-1 receptor ligand (PD-L1) and CD16 proportions, were used to identify and classify patients into one of three mutually exclusive TME classes: immune-scarce, immune-intermediate, and immune-rich tumors. Patients with immune-rich tumors were characterized by a lower proportion of melanoma cells and higher proportions of immune cells, including higher PD-L1 expression. These patients had higher response rates and longer progression-free survival (PFS) than those with immune-intermediate and immune-scarce tumors. At a median follow-up of 18 months (95% CI: 6.7 to 49 months), the 1-year PFS was 76% (95% CI: 64% to 90%) for patients with an immune-rich tumor, 56% (95% CI: 44% to 72%) for those with an immune-intermediate tumor, and 33% (95% CI: 23% to 47%) for patients with an immune-scarce tumor. A higher response rate was observed in patients with an immune-scarce or immune-intermediate tumor when treated with IPI+PD-1 compared with those treated with PD-1 alone. CONCLUSIONS: Our study provides an automatic TME classification method that may predict the clinical efficacy of immunotherapy for patients with metastatic melanoma.

Cross-platform comparison of immune signatures in immunotherapy-treated patients with advanced melanoma using a rank-based scoring approach.

BACKGROUND: Gene expression profiling is increasingly being utilised as a diagnostic, prognostic and predictive tool for managing cancer patients. Single-sample scoring approach has been developed to alleviate instability of signature scores due to variations from sample composition. However, it is a challenge to achieve comparable signature scores across different expressional platforms. METHODS: The pre-treatment biopsies from a total of 158 patients, who have received single-agent anti-PD-1 (n = 84) or anti-PD-1 + anti-CTLA-4 therapy (n = 74), were performed using NanoString PanCancer IO360 Panel. Multiple immune-related signature scores were measured from a single-sample rank-based scoring approach, singscore. We assessed the reproducibility and the performance in reporting immune profile of singscore based on NanoString assay in advance melanoma. To conduct cross-platform analyses, singscores between the immune profiles of NanoString assay and the previous orthogonal whole transcriptome sequencing (WTS) data were compared through linear regression and cross-platform prediction. RESULTS: singscore-derived signature scores reported significantly high scores in responders in multiple PD-1, MHC-1-, CD8 T-cell-, antigen presentation-, cytokine- and chemokine-related signatures. We found that singscore provided stable and reproducible signature scores among the repeats in different batches and cross-sample normalisations. The cross-platform comparisons confirmed that singscores derived via NanoString and WTS were comparable. When singscore of WTS generated by the overlapping genes to the NanoString gene set, the signatures generated highly correlated cross-platform scores (Spearman correlation interquartile range (IQR) [0.88, 0.92] and r2 IQR [0.77, 0.81]) and better prediction on cross-platform response (AUC = 86.3%). The model suggested that Tumour Inflammation Signature (TIS) and Personalised Immunotherapy Platform (PIP) PD-1 are informative signatures for predicting immunotherapy-response outcomes in advanced melanoma patients treated with anti-PD-1-based therapies. CONCLUSIONS: Overall, the outcome of this study confirms that singscore based on NanoString data is a feasible approach to produce reliable signature scores for determining patients' immune profiles and the potential clinical utility in biomarker implementation, as well as to conduct cross-platform comparisons, such as WTS.

Detailed spatial immunophenotyping of primary melanomas reveals immune cell subpopulations associated with patient outcome.

While the tumor immune microenvironment (TIME) of metastatic melanoma has been well characterized, the primary melanoma TIME is comparatively poorly understood. Additionally, although the association of tumor-infiltrating lymphocytes with primary melanoma patient outcome has been known for decades, it is not considered in the current AJCC melanoma staging system. Detailed immune phenotyping of advanced melanoma has revealed multiple immune biomarkers, including the presence of CD8+ T-cells, for predicting response to immunotherapies. However, in primary melanomas, immune biomarkers are lacking and CD8+ T-cells have yet to be extensively characterized. As recent studies combining immune features and clinicopathologic characteristics have created more accurate predictive models, this study sought to characterize the TIME of primary melanomas and identify predictors of patient outcome. We first phenotyped CD8+ T cells in fresh stage II primary melanomas using flow cytometry (n = 6), identifying a CD39+ tumor-resident CD8+ T-cell subset enriched for PD-1 expression. We then performed Opal multiplex immunohistochemistry and quantitative pathology-based immune profiling of CD8+ T-cell subsets, along with B cells, NK cells, Langerhans cells and Class I MHC expression in stage II primary melanoma specimens from patients with long-term follow-up (n = 66), comparing patients based on their recurrence status at 5 years after primary diagnosis. A CD39+CD103+PD-1- CD8+ T-cell population (P2) comprised a significantly higher proportion of intratumoral and stromal CD8+ T-cells in patients with recurrence-free survival (RFS) ≥5 years vs those with RFS <5 years (p = 0.013). Similarly, intratumoral B cells (p = 0.044) and a significantly higher B cell density at the tumor/stromal interface were associated with RFS. Both P2 and B cells localized in significantly closer proximity to melanoma cells in patients who remained recurrence-free (P2 p = 0.0139, B cell p = 0.0049). Our results highlight how characterizing the TIME in primary melanomas may provide new insights into how the complex interplay of the immune system and tumor can modify the disease outcomes. Furthermore, in the context of current clinical trials of adjuvant anti-PD-1 therapies in high-risk stage II primary melanoma, assessment of B cells and P2 could identify patients at risk of recurrence and aid in long-term treatment decisions at the point of primary melanoma diagnosis.