Automated tumor proportion score analysis for PD-L1 (22C3) expression in lung squamous cell carcinoma

Artificial Intelligence-Based Treatment Decisions: A New Era for NSCLC

Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related mortality among women and men, in developed countries, despite the public health interventions including tobacco-free campaigns, screening and early detection methods, recent therapeutic advances, and ongoing intense research on novel antineoplastic modalities. Targeting oncogenic driver mutations and immune checkpoint inhibition has indeed revolutionized NSCLC treatment, yet there still remains the unmet need for robust and standardized predictive biomarkers to accurately inform clinical decisions. Artificial intelligence (AI) represents the computer-based science concerned with large datasets for complex problem-solving. Its concept has brought a paradigm shift in oncology considering its immense potential for improved diagnosis, treatment guidance, and prognosis. In this review, we present the current state of AI-driven applications on NSCLC management, with a particular focus on radiomics and pathomics, and critically discuss both the existing limitations and future directions in this field. The thoracic oncology community should not be discouraged by the likely long road of AI implementation into daily clinical practice, as its transformative impact on personalized treatment approaches is undeniable.

Deep Learning for Lung Cancer Diagnosis, Prognosis and Prediction Using Histological and Cytological Images: A Systematic Review

Lung cancer is one of the deadliest cancers worldwide, with a high incidence rate, especially in tobacco smokers. Lung cancer accurate diagnosis is based on distinct histological patterns combined with molecular data for personalized treatment. Precise lung cancer classification from a single H&E slide can be challenging for a pathologist, requiring most of the time additional histochemical and special immunohistochemical stains for the final pathology report. According to WHO, small biopsy and cytology specimens are the available materials for about 70% of lung cancer patients with advanced-stage unresectable disease. Thus, the limited available diagnostic material necessitates its optimal management and processing for the completion of diagnosis and predictive testing according to the published guidelines. During the new era of Digital Pathology, Deep Learning offers the potential for lung cancer interpretation to assist pathologists' routine practice. Herein, we systematically review the current Artificial Intelligence-based approaches using histological and cytological images of lung cancer. Most of the published literature centered on the distinction between lung adenocarcinoma, lung squamous cell carcinoma, and small cell lung carcinoma, reflecting the realistic pathologist's routine. Furthermore, several studies developed algorithms for lung adenocarcinoma predominant architectural pattern determination, prognosis prediction, mutational status characterization, and PD-L1 expression status estimation.

The artificial intelligence and machine learning in lung cancer immunotherapy

Since the past decades, more lung cancer patients have been experiencing lasting benefits from immunotherapy. It is imperative to accurately and intelligently select appropriate patients for immunotherapy or predict the immunotherapy efficacy. In recent years, machine learning (ML)-based artificial intelligence (AI) was developed in the area of medical-industrial convergence. AI can help model and predict medical information. A growing number of studies have combined radiology, pathology, genomics, proteomics data in order to predict the expression levels of programmed death-ligand 1 (PD-L1), tumor mutation burden (TMB) and tumor microenvironment (TME) in cancer patients or predict the likelihood of immunotherapy benefits and side effects. Finally, with the advancement of AI and ML, it is believed that "digital biopsy" can replace the traditional single assessment method to benefit more cancer patients and help clinical decision-making in the future. In this review, the applications of AI in PD-L1/TMB prediction, TME prediction and lung cancer immunotherapy are discussed.

The role of Immunotherapy in esophageal and gastric cancer

Upper gastrointestinal tract tumors historically have a poor prognosis. The decision to treat esophageal or gastric cancers by surgery, radiotherapy, systemic therapy, or a combination of these treatment modalities should always be discussed multidisciplinary. The introduction of immunotherapy has drastically transformed the treatment landscape of multiple solid malignancies. Emerging data from early and late phase clinical trials suggests that the use of immunotherapies that target immune checkpoint proteins such as PD-1/PD-L1 result in superior overall survival in advanced, metastatic, or recurrent esophageal and gastric cancer, whether or not with specific molecular characteristics such as PD-L1 expression level or microsatellite instability. This review offers an overview of the most recent advances in the field of immunotherapy treatment in esophageal and gastric cancer.

Subtyping of advanced lung cancer based on PD-L1 expression, tumor histopathology and mutation burden (EGFR and KRAS): a study from North India

Immune checkpoint inhibitor (PD-L1) therapy of advanced non-small-cell lung cancer (NSCLC) has variable outcomes. Tumor subtypes based on PD-L1 expression, histopathology, mutation burden is required for patient stratification and formulation of treatment guidelines. Lung cancers (n=57) diagnosed at Pathology department, VPCI (2018-2021) were retrospectively analyzed. PD-L1(SP263) expressed by tumor cells [low (<1%), medium (1-49%), high (≥50%)] was correlated with histopathology, microenvironment, EGFR, KRAS expression. Patients were categorized into high and low risk based on their: i) gender: males (n=47, 30-89 years), females (n=10, 45-80 years); ii) smoking history: males 26/47 (45.61%), females 1/10 (10%); iii) tumor subtyping: squamous cell carcinoma 15/57 (26.32%), adenocarcinoma 6/57 (17.54%), NSCLC-undifferentiated 24/57 (42.10%), adenosquamous carcinoma 5/57 (8.77 %), carcinosarcoma 4/57 (7.02%), small cell carcinoma 1/57 (1.75%); iv) inflammatory tumor microenvironment/TILs 44/57 (77.1%); iv) PD-L1 positivity-31/57 (54.3%); v) concomitant EGFR/KRAS positivity. PD-L1positive cases showed squamous/undifferentiated histopathology, concomitant EGFR+ (9/20, 45%) and KRAS+ (8/15, 53.3%), smoking+ (21/31,67.74%).PD-L1 negative cases (26/57, 45.6%), were EGFR+ (2/14, 14.28%) and KRAS+ (6/19, 31.5%). The high-risk lung cancer subtypes show squamous/undifferentiated histopathology, inflammatory microenvironment, male preponderance, smoking history, higher concomitant PD-L1, KRAS and EGFR positivity. Lung cancer subtyping can predict clinical response/resistance of patients prior to initiation of PD-L1 inhibitor therapies and can be used to guide therapy.

Facts and Hopes on the Use of Artificial Intelligence for Predictive Immunotherapy Biomarkers in Cancer

Immunotherapy by immune checkpoint inhibitors has become a standard treatment strategy for many types of solid tumors. However, the majority of patients with cancer will not respond, and predicting response to this therapy is still a challenge. Artificial intelligence (AI) methods can extract meaningful information from complex data, such as image data. In clinical routine, radiology or histopathology images are ubiquitously available. AI has been used to predict the response to immunotherapy from radiology or histopathology images, either directly or indirectly via surrogate markers. While none of these methods are currently used in clinical routine, academic and commercial developments are pointing toward potential clinical adoption in the near future. Here, we summarize the state of the art in AI-based image biomarkers for immunotherapy response based on radiology and histopathology images. We point out limitations, caveats, and pitfalls, including biases, generalizability, and explainability, which are relevant for researchers and health care providers alike, and outline key clinical use cases of this new class of predictive biomarkers.

Artificial Intelligence-Assisted Score Analysis for Predicting the Expression of the Immunotherapy Biomarker PD-L1 in Lung Cancer

Programmed cell death ligand 1 (PD-L1) is a critical biomarker for predicting the response to immunotherapy. However, traditional quantitative evaluation of PD-L1 expression using immunohistochemistry staining remains challenging for pathologists. Here we developed a deep learning (DL)-based artificial intelligence (AI) model to automatically analyze the immunohistochemical expression of PD-L1 in lung cancer patients. A total of 1,288 patients with lung cancer were included in the study. The diagnostic ability of three different AI models (M1, M2, and M3) was assessed in both PD-L1 (22C3) and PD-L1 (SP263) assays. M2 and M3 showed improved performance in the evaluation of PD-L1 expression in the PD-L1 (22C3) assay, especially at 1% cutoff. Highly accurate performance in the PD-L1 (SP263) was also achieved, with accuracy and specificity of 96.4 and 96.8% in both M2 and M3, respectively. Moreover, the diagnostic results of these three AI-assisted models were highly consistent with those from the pathologist. Similar performances of M1, M2, and M3 in the 22C3 dataset were also obtained in lung adenocarcinoma and lung squamous cell carcinoma in both sampling methods. In conclusion, these results suggest that AI-assisted diagnostic models in PD-L1 expression are a promising tool for improving the efficiency of clinical pathologists.

Graph-Embedded Online Learning for Cell Detection and Tumour Proportion Score Estimation

Cell detection in microscopy images can provide useful clinical information. Most methods based on deep learning for cell detection are fully supervised. Without enough labelled samples, the accuracy of these methods would drop rapidly. To handle limited annotations and massive unlabelled data, semi-supervised learning methods have been developed. However, many of these are trained off-line, and are unable to process new incoming data to meet the needs of clinical diagnosis. Therefore, we propose a novel graph-embedded online learning network (GeoNet) for cell detection. It can locate and classify cells with dot annotations, saving considerable manpower. Trained by both historical data and reliable new samples, the online network can predict nuclear locations for upcoming new images while being optimized. To be more easily adapted to open data, it engages dynamic graph regularization and learns the inherent nonlinear structures of cells. Moreover, GeoNet can be applied to downstream tasks such as quantitative estimation of tumour proportion score (TPS), which is a useful indicator for lung squamous cell carcinoma treatment and prognostics. Experimental results for five large datasets with great variability in cell type and morphology validate the effectiveness and generalizability of the proposed method. For the lung squamous cell carcinoma (LUSC) dataset, the detection F1-scores of GeoNet for negative and positive tumour cells are 0.734 and 0.769, respectively, and the relative error of GeoNet for TPS estimation is 11.1%.

PD-L1 expression in breast invasive ductal carcinoma with incomplete pathological response to neoadjuvant chemotherapy

Objectives:

To investigate the expression of programmed death-ligand 1 (PD-L1) in breast cancer in association with incomplete pathological response (PR) to neoadjuvant chemotherapy (NAC).

Methods

PD-L1 expression was evaluated using immunohistochemistry in post-operative, post-NAC samples of 60 patients (n = 60) diagnosed with breast invasive ductal carcinoma with incomplete PR to NAC, including 31 matched pre-NAC and post-NAC samples (n = 31). PD-L1 protein expression was assessed using three scoring approaches, including the tumor proportion score (TPS), the immune cell score (ICS), and the combined tumor and immune cell score (combined positive score, CPS) with a 1% cut-off.

Results

In the post-operative, post-NAC samples (n = 60), positive expression rate of PD-L1 was observed in 18.3% (11/60) of cases by TPS, 31.7% (19/60) by ICS, and 25% (15/60) by CPS. In matched samples, positive expression rate of PD-L1 was observed in 19.3% (6/31) of patients by TPS, 51.6% (16/31) by ICS, and 19.3% (6/31) by CPS in pre-NAC specimens, while it was observed in 22.6% (7/31) of matched post-NAC samples by TPS, 22.6% (7/31) by ICS, and 19.3% (6/31) by CPS. In the matched samples, there was a significant decrease in PD-L1 immunoexpression using ICS in post-NAC specimens (McNemar’s, p = 0.020), while no significant differences were found using TPS and CPS between pre- and post-NAC samples (p = 1.000, p = 0.617; respectively). PD-L1 immunoexpression determined by TPS or CPS was only significantly associated with ER status (p = 0.022, p = 0.021; respectively), but not with other clinicopathological variables. We could not establish a correlation between PD-L1 expression and the overall survival rate (p > 0.05). There were no significant differences in the tumor infiltrating lymphocytes count between the paired pre- and post-NAC samples (t = 0.581, p = 0.563 or Wilcoxon’s Signed Rank test; z = -0.625, p = 0.529).

Conclusion

Our findings indicate that PD-L1 protein expression in infiltrating immune cells was significantly reduced in breast tumors that developed incomplete PR following the exposure to NAC.

PD-L1 expression and Tumor mutation burden as Pathological response biomarkers of Neoadjuvant immunotherapy for Early-stage Non-small cell lung cancer: A systematic review and meta-analysis

To date, there is no approved biomarker for predicting pathological response in neoadjuvant programmed cell death (ligand) 1 (PD-(L)1) blockades treated early-stage non-small cell lung cancer (NSCLC). Databases including PubMed, Embase, ClinicalTrials.gov, and Conference abstracts were searched for clinical trials of neoadjuvant PD-1/PD-L1 blockades for resectable NSCLC. Data regarding major pathological response (MPR), pathological complete response (pCR) in patients with high/low pretreatment PD-L1 expression, and tumor mutation burden (TMB) were synthesized using fixed-model meta-analysis and evaluated by odds ratio with 95% confidence interval. This analysis included 10 studies involving 461 NSCLC patients. Compared with PD-L1 expression <1%, PD-L1 expression ≥1% is associated with a higher rate of MPR and pCR. High-TMB associated with MPR and pCR. Similar findings were observed in subgroup analyses despite mono-PD-1/PD-L1 blockade or their combination with chemotherapy. Notably, 50% as the cutoff value for PD-L1 expression demonstrated better prediction efficacy for MPR than that of 1%.