Medical treatment of acromegaly is currently performed through a trial-error approach using first generation somatostatin receptor ligands (fgSRLs) as first-line drugs, with an effectiveness of about 50%, and subsequent drugs are indicated through clinical judgment. Some biomarkers can predict fgSRLs response. Here we report the results of the ACROFAST study, a clinical trial in which a protocol based on predictive biomarkers of fgSRLs was evaluated.
METHODS: prospective trial (21 university hospitals) comparing the effectiveness and time-to control of two treatment protocols during 12 months: A) A personalized protocol in which first option were fgSRLs as monotherapy or in combination with pegvisomant or, pegvisomant as monotherapy depending on the short Acute Octreotide Test (sAOT) results, tumor T2 Magnetic Resonance (MRI) signal or immunostaining for E-cadherin and, B) A control group with treatment always started by fgSRLs and the other drugs included after demonstrating inadequate control.
RESULTS: Eighty-five patients participated; 45 in the personalized and 40 in the control group. More patients in the personalized protocol achieved hormonal control compared to those in the control group (78% vs 53%, p < 0.05). Survival analysis revealed a hazard ratio for achieving hormonal control adjusted by age and sex of 2.53 (CI 1.30-4.80). Patients from personalized arm were controlled in a shorter period of time (p = 0.01).
CONCLUSIONS: Personalized medicine is feasible using a relatively simple protocol and allows a higher number of patients achieving control in a shorter period of time.

Multi-omics sequencing is poised to revolutionize clinical care in the coming decade. However, there is a lack of effective and interpretable genome-wide modeling methods for the rational selection of patients for personalized interventions. To address this, we present iGenSig-Rx, an integral genomic signature-based approach, as a transparent tool for modeling therapeutic response using clinical trial datasets. This method adeptly addresses challenges related to cross-dataset modeling by capitalizing on high-dimensional redundant genomic features, analogous to reinforcing building pillars with redundant steel rods. Moreover, it integrates adaptive penalization of feature redundancy on a per-sample basis to prevent score flattening and mitigate overfitting. We then developed a purpose-built R package to implement this method for modeling clinical trial datasets. When applied to genomic datasets for HER2 targeted therapies, iGenSig-Rx model demonstrates consistent and reliable predictive power across four independent clinical trials. More importantly, the iGenSig-Rx model offers the level of transparency much needed for clinical application, allowing for clear explanations as to how the predictions are produced, how the features contribute to the prediction, and what are the key underlying pathways. We anticipate that iGenSig-Rx, as an interpretable class of multi-omics modeling methods, will find broad applications in big-data based precision oncology. The R package is available: https://github.com/wangxlab/iGenSig-Rx .

Clinical outcome prediction is important for stratified therapeutics. Machine learning (ML) and deep learning (DL) methods facilitate therapeutic response prediction from transcriptomic profiles of cells and clinical samples. Clinical transcriptomic DL is challenged by the low-sample sizes (34-286 subjects), high-dimensionality (up to 21,653 genes) and unordered nature of clinical transcriptomic data. The established methods rely on ML algorithms at accuracy levels of 0.6-0.8 AUC/ACC values. Low-sample DL algorithms are needed for enhanced prediction capability. Here, an unsupervised manifold-guided algorithm was employed for restructuring transcriptomic data into ordered image-like 2D-representations, followed by efficient DL of these 2D-representations with deep ConvNets. Our DL models significantly outperformed the state-of-the-art (SOTA) ML models on 82% of 17 low-sample benchmark datasets (53% with >0.05 AUC/ACC improvement). They are more robust than the SOTA models in cross-cohort prediction tasks, and in identifying robust biomarkers and response-dependent variational patterns consistent with experimental indications.

Background: Glioma patients often experience unfavorable outcomes and elevated mortality rates. Our study established a prognostic signature utilizing cuproptosis-associated long non-coding RNAs (CRLs) and identified novel prognostic biomarkers and therapeutic targets for glioma. Methods: The expression profiles and related data of glioma patients were obtained from The Cancer Genome Atlas, an accessible online database. We then constructed a prognostic signature using CRLs and evaluated the prognosis of glioma patients by means of Kaplan-Meier survival curves and receiver operating characteristic curves. A nomogram based on clinical features was employed to predict the individual survival probability of glioma patients. Functional enrichment analysis was conducted to identify crucial CRL-related enriched biological pathways. The role of LEF1-AS1 in glioma was validated in two glioma cell lines (T98 and U251). Results: We developed and validated a prognostic model for glioma with 9 CRLs. Patients with low-risk had a considerably longer overall survival (OS). The prognostic CRL signature may serve independently as an indicator of prognosis for glioma patients. In addition, functional enrichment analysis revealed significant enrichment of multiple immunological pathways. Notable differences were observed between the two risk groups in terms of immune cell infiltration, function, and immune checkpoints. We further identified four drugs based on their different IC50 values from the two risk groups. Subsequently, we discovered two molecular subtypes of glioma (cluster one and cluster two), with the cluster one subtype exhibiting a remarkably longer OS compared to the cluster two subtype. Finally, we observed that inhibition of LEF1-AS1 curbed the proliferation, migration, and invasion of glioma cells. Conclusion: The CRL signatures were confirmed as a reliable prognostic and therapy response indicator for glioma patients. Inhibition of LEF1-AS1 effectively suppressed the growth, migration, and invasion of gliomas; therefore, LEF1-AS1 presents itself as a promising prognostic biomarker and potential therapeutic target for glioma.

Pancreatic cancer is one of the most lethal tumors owing to its unspecific symptoms during the early stage and multiple treatment resistances. Pyroptosis, a newly discovered gasdermin-mediated cell death, facilitates anti- or pro-tumor effects in a variety of cancers, whereas the impact of pyroptosis in pancreatic cancer remains unclear. Therefore, we downloaded RNA expression and clinic data from the TCGA-PAAD cohort and were surprised to find that most pyroptosis-related genes (PRGs) are not only overexpressed in tumor tissue but also strongly associated with overall survival. For their remarkable prognostic value, cox regression analysis and lasso regression were used to establish a five-gene signature. All patients were divided into low- and high-risk groups based on the media value of the risk score, and we discovered that low-risk patients had better outcomes in both the testing and validation cohorts using time receiver operating characteristic (ROC), nomograms, survival, and decision analysis. More importantly, a higher somatic mutation burden and less immune cell infiltration were found in the high-risk group. Following that, we predicted tumor response to chemotherapy and immunotherapy in both low- and high-risk groups, which suggests patients with low risk were more likely to respond to both immunotherapy and chemotherapy. To summarize, our study established an effective model that can help clinicians better predict patients\' drug responses and outcomes, and we also present basic evidence for future pyroptosis related studies in pancreatic cancer.

Angiogenesis is a common feature of many physiological processes and pathological conditions. RGD-containing peptides can strongly bind to integrin αvβ3 expressed on endothelial cells in neovessels and several tumor cells with high specificity, making them promising molecular agents for imaging angiogenesis. Although studies of RGD-containing peptides combined with radionuclides, namely, 18F, 64Cu, and 68Ga for positron emission tomography (PET) imaging have shown high spatial resolution and accurate quantification of tracer uptake, only a few of these radiotracers have been successfully translated into clinical use. This review summarizes the RGD-based tracers in terms of accumulation in tumors and adjacent tissues, and comparison with traditional 18F-fluorodeoxyglucose (FDG) imaging. The value of RGD-based tracers for diagnosis, differential diagnosis, tumor subvolume delineation, and therapeutic response prediction is mainly discussed. Very low RGD accumulation, in contrast to high FDG metabolism, was found in normal brain tissue, indicating that RGD-based imaging provides an excellent tumor-to-background ratio for improved brain tumor imaging. However, the intensity of the RGD-based tracers is much higher than FDG in normal liver tissue, which could lead to underestimation of primary or metastatic lesions in liver. In multiple studies, RGD-based imaging successfully realized the diagnosis and differential diagnosis of solid tumors and also the prediction of chemoradiotherapy response, providing complementary rather than similar information relative to FDG imaging. Of most interest, baseline RGD uptake values can not only be used to predict the tumor efficacy of antiangiogenic therapy, but also to monitor the occurrence of adverse events in normal organs. This unique dual predictive value in antiangiogenic therapy may be better than that of FDG-based imaging.

Immune checkpoint therapies such as PD-1 blockade have vastly improved the treatment of numerous cancers, including basal cell carcinoma (BCC). However, patients afflicted with pancreatic ductal carcinoma (PDAC), one of the deadliest malignancies, overwhelmingly exhibit negative responses to checkpoint therapy. We sought to combine data analysis and machine learning to differentiate the putative mechanisms of BCC and PDAC non-response. We discover that increased MHC-I expression in malignant cells and suppression of MHC and PD-1/PD-L expression in CD8+ T cells is associated with nonresponse to treatment. Furthermore, we leverage machine learning to predict response to PD-1 blockade on a cellular level. We confirm divergent resistance mechanisms between BCC, PDAC, and melanoma and highlight the potential for rapid and affordable testing of gene expression in BCC patients to accurately predict response to checkpoint therapies. Our findings present an optimistic outlook for the use of quantitative cross-cancer analyses in characterizing immune responses and predicting immunotherapy outcomes.

Metabolic reprogramming contributes to patient prognosis. Here, we aimed to reveal the comprehensive landscape in metabolism of head and neck squamous carcinoma (HNSCC), and establish a novel metabolism-related prognostic model to explore the clinical potential and predictive value on therapeutic response. We screened 4752 metabolism-related genes (MRGs) and then identified differentially expressed MRGs in HNSCC. A novel 10-MRGs risk model for prognosis was established by the univariate Cox regression analysis and the least absolute shrinkage and selection operator (Lasso) regression analysis, and then verified in both internal and external validation cohort. Kaplan-Meier analysis was employed to explore its prognostic power on the response of conventional therapy. The immune cell infiltration was also evaluated and we used tumor immune dysfunction and exclusion (TIDE) algorithm to estimate potential response of immunotherapy in different risk groups. Nomogram model was constructed to further predict patients\' prognoses. We found the MRGs-related prognostic model showed good prediction performance. Survival analysis indicated that patients suffered obviously poorer survival outcomes in high-risk group (p < 0.001). The metabolism-related signature was further confirmed to be the independent prognostic value of HNSCC (HR = 6.387, 95% CI = 3.281-12.432, p < 0.001), the efficacy of predictive model was also verified by internal and external validation cohorts. We observed that HNSCC patients would benefit from the application of chemotherapy in the low-risk group (p = 0.029). Immunotherapy may be effective for HNSCC patients with high risk score (p < 0.01). Furthermore, we established a predictive nomogram model for clinical application with high performance. Our study constructed and validated a promising 10-MRGs signature for monitoring outcome, which may provide potential indicators for metabolic therapy and therapeutic response prediction in HNSCC.