From pixels to patient care: deep learning-enabled pathomics signature offers precise outcome predictions for immunotherapy in esophageal squamous cell cancer.

BACKGROUND: Immunotherapy has significantly improved survival of esophageal squamous cell cancer (ESCC) patients, however the clinical benefit was limited to only a small portion of patients. This study aimed to perform a deep learning signature based on H&E-stained pathological specimens to accurately predict the clinical benefit of PD-1 inhibitors in ESCC patients. METHODS: ESCC patients receiving PD-1 inhibitors from Shandong Cancer Hospital were included. WSI images of H&E-stained histological specimens of included patients were collected, and randomly divided into training (70%) and validation (30%) sets. The labels of images were defined by the progression-free survival (PFS) with the interval of 4 months. The pretrained ViT model was used for patch-level model training, and all patches were projected into probabilities after linear classifier. Then the most predictive patches were passed to RNN for final patient-level prediction to construct ESCC-pathomics signature (ESCC-PS). Accuracy rate and survival analysis were performed to evaluate the performance of ViT-RNN survival model in validation cohort. RESULTS: 163 ESCC patients receiving PD-1 inhibitors were included for model training. There were 486,188 patches of 1024*1024 pixels from 324 WSI images of H&E-stained histological specimens after image pre-processing. There were 120 patients with 227 images in training cohort and 43 patients with 97 images in validation cohort, with balanced baseline characteristics between two groups. The ESCC-PS achieved an accuracy of 84.5% in the validation cohort, and could distinguish patients into three risk groups with the median PFS of 2.6, 4.5 and 12.9 months (P < 0.001). The multivariate cox analysis revealed ESCC-PS could act as an independent predictor of survival from PD-1 inhibitors (P < 0.001). A combined signature incorporating ESCC-PS and expression of PD-L1 shows significantly improved accuracy in outcome prediction of PD-1 inhibitors compared to ESCC-PS and PD-L1 anlone, with the area under curve value of 0.904, 0.924, 0.610 for 6-month PFS and C-index of 0.814, 0.806, 0.601, respectively. CONCLUSIONS: The outcome supervised pathomics signature based on deep learning has the potential to enable superior prognostic stratification of ESCC patients receiving PD-1 inhibitors, which convert the images pixels to an effective and labour-saving tool to optimize clinical management of ESCC patients.

Outcome-Supervised Deep Learning on Pathologic Whole Slide Images for Survival Prediction of Immunotherapy in Patients with Non-Small Cell Lung Cancer.

Although programmed death-(ligand) 1 (PD-(L)1) inhibitors are marked by durable efficacy in patients with non-small cell lung cancer (NSCLC), approximately 60% of the patients still suffer from recurrence and metastasis after PD-(L)1 inhibitor treatment. To accurately predict the response to PD-(L)1 inhibitors, we presented a deep learning model using a Vision Transformer (ViT) network based on hematoxylin and eosin (H&E)-stained specimens of patients with NSCLC. Two independent cohorts of patients with NSCLC receiving PD-(L)1 inhibitors from Shandong Cancer Hospital and Institute and Shandong Provincial Hospital were enrolled for model training and external validation, respectively. Whole slide images (WSIs) of H&E-stained histologic specimens were obtained from these patients and patched into 1024 × 1024 pixels. The patch-level model was trained based on ViT to identify the predictive patches, and patch-level probability distribution was performed. Then, we trained a patient-level survival model based on the ViT-Recursive Neural Network framework and externally validated it in the Shandong Provincial Hospital cohort. A total of 291 WSIs of H&E-stained histologic specimens from 198 patients with NSCLC in Shandong Cancer Hospital and 62 WSIs from 30 patients with NSCLC in Shandong Provincial Hospital were included in the model training and validation. The model achieved an accuracy of 88.6% in the internal validation cohort and 81% in the external validation cohort. The survival model also remained a statistically independent predictor of survival from PD-(L)1 inhibitors. In conclusion, the outcome-supervised ViT-Recursive Neural Network survival model based on pathologic WSIs could be used to predict immunotherapy efficacy in patients with NSCLC.

East Asian patients who received immunotherapy-based therapy associated with improved survival benefit in advanced non-small cell lung cancer: An updated meta-analysis.

BACKGROUND: Immune checkpoint inhibitors (ICIs) combined with chemotherapy have been recommended as the standard treatment for advanced NSCLC patients without driver-gene mutations. However, there are different genetic characteristics and biological traits of tumors between non-East Asian (nEA) and East Asian (EA) patients with NSCLC, which may contribute to differences in the efficacy of ICIs in different ethnic populations. Previous findings regarding differences in the efficacy of ICIs among ethnic groups have been inconsistent. Therefore, we performed a meta-analysis by collecting published data to investigate the clinical outcomes of ICIs for EA NSCLC patients compared to nEA patients. METHODS: Overall survival (OS) and progression-free survival (PFS) were used to access the difference in survival outcomes between the two populations. Subgroup analyses were performed based on the line of ICIs, the use of ICIs alone or in combination, and the type of ICIs. RESULTS: A total of 9826 NSCLC patients from 21 randomized controlled trials (RCTs) with 4064 EAs were included, which involved PD-1, PD-L1, and CTLA-4 inhibitors. EA NSCLC patients who received ICIs-based therapy were associated with significantly improved survival benefits in OS (p = 0.02) compared with nEA patients. Subgroup analysis indicated that EA patients receiving first-line ICIs showed significantly superior OS compared with nEA patients (p = 0.007). Chemo-ICIs treatment showed significant advantages in terms of OS (p = 0.002) and PFS (p = 0.02) among EA patients compared to nEA patients. In addition, PD-1 inhibitors were associated with improved OS among both EA patients and nEA patients compared with PD-L1 inhibitors. CONCLUSION: EA NSCLC patients who received ICIs-based therapy were associated with significantly improved survival benefits compared with nEA NSCLC patients. Earlier intervention with ICIs and combination treatment was more recommended for EA NSCLC patients. Moreover, PD-1 inhibitors are associated with prolonged survival among both EA and nEA patients.

Efficacy and safety of thoracic radiotherapy in extensive-stage small-cell lung cancer patients receiving first-line immunotherapy plus chemotherapy: a propensity score matched multicentre retrospective analysis.

BACKGROUND: Platinum-etoposide chemotherapy combined with immune checkpoint inhibitors (ICIs) has been recommended as the first-line standard treatment for extensive-stage small-cell lung cancer (ES-SCLC). However, the effect of thoracic radiotherapy (TRT) on these patients is still unknown. This study aimed to evaluate the efficacy and safety of TRT for ES-SCLC patients who responded to first-line ICIs and chemotherapy (CHT). METHODS: Patients who received 4 to 6 cycles of ICIs and CHT as first-line therapy at three hospitals between 2018 and 2022 were included in the analysis. All patients were divided into two groups based on whether they received TRT as first-line treatment, and propensity score matching (PSM) was performed to ensure that the characteristics of two groups were well-balanced. The primary endpoints were overall survival (OS) and progression-free survival (PFS), and the secondary endpoint was toxic effects. RESULTS: A total of 276 patients were included, and the median follow-up time was 22.3 (range, 4.0-53.73) months. After PSM, 197 patients were further analysed, and 99 of whom received TRT. The baseline characteristics were well-balanced between patients in the TRT and non-TRT groups. There were significant differences in PFS between the TRT and non-TRT groups, with the median PFS of 10.76 and 7.63 months, respectively (P = 0.014). Significantly improved OS was observed in the TRT group (21.67 vs. 16.6 months, P = 0.009). In addition, the use of TRT was an independent prognostic factor for PFS and OS of ES-SCLC patients receiving ICIs plus CHT. In terms of safety, no significant increase of any grades adverse event (AE) (P = 0.874) and G3-4 AE (P = 0.909) was observed for patients receiving TRT. Radiation esophagitis, gastrointestinal and hematologic toxicities were the most common AEs in TRT group, which were tolerable. And high-dose radiotherapy was associated with higher incidence of pneumonitis. CONCLUSION: Addition of TRT showed significant survival benefits and well tolerability in ES-SCLC patients receiving platinum-etoposide CHT and ICIs, which could be a feasible first-line treatment strategy for ES-SCLC patients.

Incidence of pneumonitis following the use of different anaplastic lymphoma kinase tyrosine kinase inhibitor regimens: An updated systematic review and meta-analysis.

OBJECTIVES: Anaplastic lymphoma kinase tyrosine kinase inhibitors (ALK TKIs) have shown remarkable clinical activity in patients with non-small-cell lung cancer (NSCLC). However, pneumonitis is a serious side effect of ALK TKIs in NSCLC patients. In this meta-analysis, we aimed to determine the incidence of ALK-TKI-associated pneumonitis. MATERIALS AND METHODS: We searched electronic databases to identify relevant studies published until August 2022. The incidence of pneumonitis was calculated using a fixed-effects model when no substantial heterogeneity was observed. Otherwise, a random-effects model was used. Subgroup analyses of different treatment groups were performed. Statistical analyses were conducted using STATA 17.0. RESULTS: Twenty-six clinical trials involving 4752 patients were eligible for analysis. All-grade pneumonitis incidence was 2.92% (95% confidence interval [CI]: 1.79%-4.27%), high-grade (Grade 3-4) pneumonitis incidence was 1.42% (95% CI: 0.84%-2.12%) and Grade 5 pneumonitis incidence was 0.09% (95% CI: 0.00%-0.28%). The subgroup analysis showed that brigatinib was associated with the highest incidence of both all-grade and high-grade pneumonitis (7.09% and 3.06%, respectively). ALK TKI treatment after chemotherapy was associated with a higher incidence of all-grade and high-grade pneumonitis than first-line ALK TKI treatment (7.73% vs. 2.26% and 3.64% vs. 1.26%, respectively). Cohorts from Japanese trials had a higher incidence of all-grade and high-grade pneumonitis. CONCLUSION: Our study provides precise data on the incidence of pneumonitis in patients receiving treatment with ALK TKIs. Overall, ALK TKIs have tolerable pulmonary toxicity. Early pneumonitis identification and treatment are required to prevent further deterioration in patients receiving treatment with brigatinib and in those who received prior chemotherapy, particularly in the Japanese population.

Integrated multi-dimensional deep neural network model improves prognosis prediction of advanced NSCLC patients receiving bevacizumab.

Background: The addition of bevacizumab was found to be associated with prolonged survival whether in combination with chemotherapy, tyrosine kinase inhibitors or immune checkpoint inhibitors in the treatment landscape of advanced non-small cell lung cancer (NSCLC) patients. However, the biomarkers for efficacy of bevacizumab were still largely unknown. This study aimed to develop a deep learning model to provide individual assessment of survival in advanced NSCLC patients receiving bevacizumab. Methods: All data were retrospectively collected from a cohort of 272 radiological and pathological proven advanced non-squamous NSCLC patients. A novel multi-dimensional deep neural network (DNN) models were trained based on clinicopathological, inflammatory and radiomics features using DeepSurv and N-MTLR algorithm. And concordance index (C-index) and bier score was used to demonstrate the discriminatory and predictive capacity of the model. Results: The integration of clinicopathologic, inflammatory and radiomics features representation was performed using DeepSurv and N-MTLR with the C-index of 0.712 and 0.701 in testing cohort. And Cox proportional hazard (CPH) and random survival forest (RSF) models were also developed after data pre-processing and feature selection with the C-index of 0.665 and 0.679 respectively. DeepSurv prognostic model, indicated with best performance, was used for individual prognosis prediction. And patients divided in high-risk group were significantly associated with inferior PFS (median PFS: 5.4 vs 13.1 months, P<0.0001) and OS (median OS: 16.4 vs 21.3 months, P<0.0001). Conclusions: The integration of clinicopathologic, inflammatory and radiomics features representation based on DeepSurv model exhibited superior predictive accuracy as non-invasive method to assist in patients counseling and guidance of optimal treatment strategies.