Deep Learning to Estimate RECIST in Patients with NSCLC Treated with PD-1 Blockade.
Ontology highlight
ABSTRACT:
SUBMITTER: Arbour KC
PROVIDER: S-EPMC7981277 | biostudies-literature |
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:Different biomarkers based on genomics variants have been used to predict the response of patients treated with PD-1/programmed death receptor 1 ligand (PD-L1) blockade. We aimed to use deep-learning algorithm to estimate clinical benefit in patients with non-small-cell lung cancer (NSCLC) before immunotherapy. Peripheral blood samples or tumor tissues of 915 patients from three independent centers were profiled by whole-exome sequencing or next-generation sequencing. Based on convolutional neural network (CNN) and three conventional machine learning (cML) methods, we used multi-panels to train the models for predicting the durable clinical benefit (DCB) and combined them to develop a nomogram model for predicting prognosis. In the three cohorts, the CNN achieved the highest area under the curve of predicting DCB among cML, PD-L1 expression, and tumor mutational burden (area under the curve [AUC] = 0.965, 95% confidence interval [CI]: 0.949-0.978, P< 0.001; AUC =0.965, 95% CI: 0.940-0.989, P< 0.001; AUC = 0.959, 95% CI: 0.942-0.976, P< 0.001, respectively). Patients with CNN-high had longer progression-free survival (PFS) and overall survival (OS) than patients with CNN-low in the three cohorts. Subgroup analysis confirmed the efficient predictive ability of CNN. Combining three cML methods (CNN, SVM, and RF) yielded a robust comprehensive nomogram for predicting PFS and OS in the three cohorts (each P< 0.001). The proposed deep-learning method based on mutational genes revealed the potential value of clinical benefit prediction in patients with NSCLC and provides novel insights for combined machine learning in PD-1/PD-L1 blockade.
| S-EPMC9677530 | biostudies-literature
Project description:Several clinical trials have shown the safety and effectiveness of PD-1/PD-L1 inhibitors in neoadjuvant therapy in resectable non-small cell lung cancer (NSCLC). However, 18-83% patients can benefit from it. In this study, we aimed to assess the association of PD-L1 expression, tumor mutation burden, copy number alteration (CNA, including copy number gain and loss) burden with the pathologic response to neoadjuvant PD-1 blockade and investigate the changes in the tumor immune microenvironment (TIME) during neoadjuvant immunotherapy in NSCLC. Pre-immunotherapy treatment tumor samples from twenty-nine NSCLC patients who received neoadjuvant immunotherapy with sintilimab, an anti-PD-1 drug, were subjected to targeted DNA sequencing and PD-L1 immunochemistry staining. The pathological response was positively correlated with tumor proportion score (TPS) of PD-L1 and negatively correlated with copy number gain (CNgain) burden. Of note, the combination of CNgain burden and TPS can better stratify major pathological response (MPR) patients than did CNgain or TPS alone. Whereas, TMB showed a limited correlation with pathological regression. Additionally, PD-1 blockade led to an increase in CD8+PD-1-T cells which was clinically relevant to MPR as evaluated by multiplex immunofluorescence. A significant reduction in CD19+ cells was observed in the Non-MPR group but not in the MPR group, indicating the involvement of B cells in improving neoadjuvant immunotherapy response in NSCLC. Together, our study provides new data for the correlation of PD-L1 expression and genomic factors with drug response in neoadjuvant immunotherapy settings in NSCLC. The changes of TIME may provide novel insight into the immune responses to neoadjuvant anti-PD-1 therapy.
| S-EPMC8758728 | biostudies-literature
Project description:Purpose: To investigate the link between treatment with CTLA-4 and PD-1 checkpoint blockade inhibitors and the development of noninfectious uveitis.Methods: A survey was distributed to uveitis specialists to identify patients who developed uveitis while receiving either PD-1 inhibitors pembrolizumab and nivolumab; PD-L1 inhibitors atezolizumab, avelumab, and durvalumab; or the CTLA-4 inhibitor ipilimumab.Results: Fifteen patients from seven institutions were identified. The most common cancer diagnosis (13/15) was malignant melanoma. Fourteen patients had a new uveitis diagnosis following checkpoint blockade administration (six anterior uveitis, six panuveitis, one posterior uveitis, one anterior/intermediate combined); one patient developed optic neuritis. Uveitis was diagnosed within 6 months after drug initiation for 11/12 patients (median 63 days). Corticosteroid treatment was effective for most patients, although two patients had permanent loss of vision.Conclusions: Patients on checkpoint inhibitor therapy should be educated to seek care if they develop ocular symptoms, and prompt referral to specialists should be incorporated into oncology protocols.
| S-EPMC6832811 | biostudies-literature
Project description:Checkpoint inhibitors have been approved for the treatment of non-small cell lung cancer (NSCLC). However, only a minority of patients demonstrate a durable clinical response. PD-L1 scoring is currently the only biomarker measure routinely used to select patients for immunotherapy, but its predictive accuracy is modest. The aim of our study was to evaluate a proteomic assay for the analysis of patient plasma in the context of immunotherapy. Pretreatment plasma samples from 43 NSCLC patients who received anti-PD-(L)1 therapy were analyzed using a proximity extension assay (PEA) to quantify 92 different immune oncology-related proteins. The plasma protein levels were associated with clinical and histopathological parameters, as well as therapy response and survival. Unsupervised hierarchical cluster analysis revealed two patient groups with distinct protein profiles associated with high and low immune protein levels, designated as "hot" and "cold". Further supervised cluster analysis based on T-cell activation markers showed that higher levels of T-cell activation markers were associated with longer progression-free survival (PFS) (p < 0.01). The analysis of single proteins revealed that high plasma levels of CXCL9 and CXCL10 and low ADA levels were associated with better response and prolonged PFS (p < 0.05). Moreover, in an explorative response prediction model, the combination of protein markers (CXCL9, CXCL10, IL-15, CASP8, and ADA) resulted in higher accuracy in predicting response than tumor PD-L1 expression or each protein assayed individually. Our findings demonstrate a proof of concept for the use of multiplex plasma protein levels as a tool for anti-PD-(L)1 response prediction in NSCLC. Additionally, we identified protein signatures that could predict the response to anti-PD-(L)1 therapy.
| S-EPMC8268315 | biostudies-literature
Project description:Advanced NSCLC harboring epidermal growth factor receptor (EGFR) mutations treated with EGFR tyrosine kinase inhibitors (TKIs) typically progresses after initial response due to acquired resistance. TKIs are often continued beyond progressive disease by RECIST. We investigated the practice of continuing EGFR-TKIs after RECIST-PD via CT findings.Among 101 advanced NSCLC patients with sensitizing EGFR mutations treated with first-line EGFR-TKIs, 70 patients had baseline and at least one follow-up CT for retrospective radiographic assessments using RECIST1.1; 56 patients had experienced PD by the data closure date of June 2011.Among 56 patients experiencing PD, 82% were female, median age was 63 years, 50% were never-smokers, 57% had distant metastasis, 57% had exon 19 deletion, and 89% were treated with erlotinib. 49 patients (88%) continued TKI therapy beyond retrospectively assessed PD. 31/32 (97%) patients who progressed by an increase in their target lesions continued TKI. 13/16 (81%) patients who progressed by appearance of a new lesion remained on TKI. 5/6 (83%) patients with both increase of target lesions and new lesion at PD continued TKI. Two patients with PD in non-target lesions discontinued therapy at PD. In 49 continuing patients, the median time from retrospectively assessed RECIST-PD to termination of TKI was 10.1 months.88% of EFGR-mutant NSCLC patients who progressed on first-line TKI continued therapy beyond RECIST-PD, which is not the single determining factor for terminating TKI in EGFR-mutant NSCLC patients. Additional radiographically defined progression criteria are needed for this population.
| S-EPMC3580850 | biostudies-literature
Project description:The interaction between programmed cell death protein 1 (PD-1) and its ligand programmed death-ligand 1 (PD-L1) induces exhaustions of cytotoxic lymphocytes in the tumor microenvironment, which facilitates tumor immune evasion. PD-1/PD-L1 blockade therapy, which prevents the receptors and ligands from binding to each other, disrupts the T-cell exhaustion signaling, thereby increasing antitumor immunity. Inspiringly, it has revolutionized the treatment of many different types of cancers including non-small-cell lung carcinoma, melanoma, lymphoma, and so on. However, with the intention of generating an antitumor immune response, PD-1/PD-L1 blockade may also lead to a spectrum of side effects. The profile of adverse events (AEs) of PD-1/PD-L1 blockade is not exactly the same with other immune checkpoint blockades, such as blockade of cytotoxic T-lymphocyte-associated protein 4. Although cutaneous, gastrointestinal, and pulmonary systems are common victims, AEs of PD-1/PD-L1 blockade might occur in any other organ system of the human body. These toxicities can be life-threatening if not managed promptly, and proper treatment intervention is imperative for optimal control and prevention of severe damage. Currently, clinical practice for the management of AEs in PD-1/PD-L1 blockade remains sporadic and variable. The majority of initial clinical trials were carried out in Caucasians. The trials of multiple races usually included a small portion of Asian participants, and results were calculated and interpreted for the entire included subjects without any race-specific conclusions. Therefore, the information on PD-1/PD-L1 blockade in Asians is far from systematic or comprehensive. Recently, as the results of clinical trials of anti-PD-1/PD-L1 agents in Asian populations have been gradually released, we summarized current evidence with a specific focus on the Asian population, hoping to outline strategies and offer guidance on the management of AEs in cancer patients treated with PD-1/PD-L1 blockade in the Asian world.
| S-EPMC6614522 | biostudies-literature
Project description:ObjectivesThis study aims to establish nomograms to accurately predict the overall survival (OS) and progression-free survival (PFS) in patients with non-small cell lung cancer (NSCLC) who received chemotherapy alone as the first-line treatment.Materials and methodsIn a training cohort of 121 NSCLC patients, radiomic features were extracted, selected from intra- and peri-tumoral regions, and used to build signatures (S1 and S2) using a Cox regression model. Deep learning features were obtained from three convolutional neural networks and utilized to build signatures (S3, S4, and S5) that were stratified into over- and under-expression subgroups for survival risk using X-tile. After univariate and multivariate Cox regression analyses, a nomogram incorporating the tumor, node, and metastasis (TNM) stages, radiomic signature, and deep learning signature was established to predict OS and PFS, respectively. The performance was validated using an independent cohort (61 patients).ResultsTNM stages, S2 and S3 were identified as the significant prognosis factors for both OS and PFS; S2 (OS: (HR (95%), 2.26 (1.40-3.67); PFS: (HR (95%), 2.23 (1.36-3.65)) demonstrated the best ability in discriminating patients with over- and under-expression. For the OS nomogram, the C-index (95% CI) was 0.74 (0.70-0.79) and 0.72 (0.67-0.78) in the training and validation cohorts, respectively; for the PFS nomogram, the C-index (95% CI) was 0.71 (0.68-0.81) and 0.72 (0.66-0.79). The calibration curves for the 3- and 5-year OS and PFS were in acceptable agreement between the predicted and observed survival. The established nomogram presented a higher overall net benefit than the TNM stage for predicting both OS and PFS.ConclusionBy integrating the TNM stage, CT radiomic signature, and deep learning signatures, the established nomograms can predict the individual prognosis of NSCLC patients who received chemotherapy. The integrated nomogram has the potential to improve the individualized treatment and precise management of NSCLC patients.
| S-EPMC10590525 | biostudies-literature
Project description:Biological collagenous tissues comprised of networks of collagen fibers are suitable for a broad spectrum of medical applications owing to their attractive mechanical properties. In this study, we developed a noninvasive approach to estimate collagenous tissue elastic properties directly from microscopy images using Machine Learning (ML) techniques. Glutaraldehyde-treated bovine pericardium (GLBP) tissue, widely used in the fabrication of bioprosthetic heart valves and vascular patches, was chosen to develop a representative application. A Deep Learning model was designed and trained to process second harmonic generation (SHG) images of collagen networks in GLBP tissue samples, and directly predict the tissue elastic mechanical properties. The trained model is capable of identifying the overall tissue stiffness with a classification accuracy of 84%, and predicting the nonlinear anisotropic stress-strain curves with average regression errors of 0.021 and 0.031. Thus, this study demonstrates the feasibility and great potential of using the Deep Learning approach for fast and noninvasive assessment of collagenous tissue elastic properties from microstructural images.Statement of significanceIn this study, we developed, to our best knowledge, the first Deep Learning-based approach to estimate the elastic properties of collagenous tissues directly from noninvasive second harmonic generation images. The success of this study holds promise for the use of Machine Learning techniques to noninvasively and efficiently estimate the mechanical properties of many structure-based biological materials, and it also enables many potential applications such as serving as a quality control tool to select tissue for the manufacturing of medical devices (e.g. bioprosthetic heart valves).
| S-EPMC5653437 | biostudies-literature
Project description:PurposeNew therapies are needed to treat immune checkpoint inhibitor-resistant non-small cell lung cancer (NSCLC) and identify biomarkers to personalize treatment. Epigenetic therapies, including histone deacetylase inhibitors, may synergize with programmed cell death-1 (PD-1) blockade to overcome resistance. We report outcomes in patients with anti-programmed cell death ligand-1 [PD-(L)1]-resistant/refractory NSCLC treated with pembrolizumab plus entinostat in ENCORE 601.Patients and methodsThe expansion cohort of ENCORE 601 included patients with NSCLC who previously experienced disease progression with immune checkpoint inhibitors. The primary endpoint for the phase II expansion cohort is overall response rate (ORR); safety, tolerability, and exploratory endpoints are described.ResultsOf 76 treated patients, 71 were evaluable for efficacy. immune-regulated RECIST-assessed ORR was 9.2% [95% confidence interval (CI): 3.8-18.1], which did not meet the prespecified threshold for positivity. Median duration of response was 10.1 months (95% CI: 3.9-not estimable), progression-free survival (PFS) at 6 months was 22%, median PFS was 2.8 months (95% CI: 1.5-4.1), and median overall survival was 11.7 months (95% CI: 7.6-13.4). Benefit was enriched among patients with high levels of circulating classical monocytes at baseline. Baseline tumor PD-L1 expression and IFNγ gene expression were not associated with benefit. Treatment-related grade ≥3 adverse events occurred in 41% of patients.ConclusionsIn anti-PD-(L)1-experienced patients with NSCLC, entinostat plus pembrolizumab did not achieve the primary response rate endpoint but provided a clinically meaningful benefit, with objective response in 9% of patients. No new toxicities, including immune-related adverse events, were seen for either drug. Future studies will continue to evaluate the association of monocyte levels and response.
| S-EPMC7887114 | biostudies-literature
Project description:BackgroundAnti-PD-1/PD-L1 drugs are effective as monotherapy in a proportion of NSCLC patients and there is a strong rationale for combining them with targeted therapy. Inhibition of MAPK pathway may have pleiotropic effects on the microenvironment. This work investigates the efficacy of combining MEK and PD-L1 inhibition in pre-clinical and ex-vivo NSCLC models.MethodsWe studied the effects of MEK inhibitors (MEK-I) on PD-L1 and MCH-I protein expression and cytokine production in vitro in NSCLC cell lines and in PBMCs from healthy donors and NSCLC patients, the efficacy of combining MEK-I with anti-PD-L1 antibody in ex-vivo human spheroid cultures obtained from fresh biopsies from NSCLC patients in terms of cell growth arrest, cytokine production and T-cell activation by flow cytometry.ResultsMEK-I modulates in-vitro the immune micro-environment through a transcriptionally decrease of PD-L1 expression, enhance of MHC-I expression on tumor cells, increase of the production of several cytokines, like IFNγ, IL-6, IL-1β and TNFα. These effects trigger a more permissive anti-tumor immune reaction, recruiting immune cells to the tumor sites. We confirmed these data on ex-vivo human spheroids, showing a synergism of MEK and PD-L1 inhibition as result of both direct cancer cell toxicity of MEK-I and its immune-stimulatory effect on cytokine secretion profile of cancer cells and PBMCs with the induction of the ones that sustain an immune-reactive and inflammatory micro-environment.ConclusionsOur work shows the biological rationale for combining immunotherapy with MEK-I in a reproducible ex-vivo 3D-culture model, useful to predict sensitivity of patients to such therapies.
| S-EPMC6567578 | biostudies-literature