Deep learning-based scoring of tumour-infiltrating lymphocytes is prognostic in primary melanoma and predictive to PD-1 checkpoint inhibition in melanoma metastases

Tumor-infiltrating lymphocytes as a prognostic and predictive factor for Melanoma

INTRODUCTION

Tumor-infiltrating lymphocytes (TILs) have been investigated as prognostic factors in melanoma. Recent advancements in assessing the tumor microenvironment in the setting of more widespread use of immune checkpoint blockade have reignited interest in identifying predictive biomarkers. This review examines the function and significance of TILs in cutaneous melanoma, evaluating their potential as prognostic and predictive markers.

AREAS COVERED

A literature search was conducted on papers covering tumor infiltrating lymphocytes in cutaneous melanoma available online in PubMed and Web of Science from inception to 1 December 2023, supplemented by citation searching. This article encompasses the assessment of TILs, the role of TILs in the immune microenvironment, TILs as a prognostic factor, TILs as a predictive factor for immunotherapy response, and clinical applications of TILs in the treatment of cutaneous melanoma.

EXPERT OPINION

Tumor-infiltrating lymphocytes play a heterogeneous role in cutaneous melanoma. While they have historically been associated with improved survival, their status as independent prognostic or predictive factors remains uncertain. Novel methods of TIL assessment, such as determination of TIL subtypes and molecular signaling, demonstrate potential for predicting therapeutic response. Further, while their clinical utility in risk-stratification in melanoma treatment shows promise, a lack of consensus data hinders standardized application.

Immune subtyping of melanoma whole slide images using multiple instance learning

Determining early-stage prognostic markers and stratifying patients for effective treatment are two key challenges for improving outcomes for melanoma patients. Previous studies have used tumour transcriptome data to stratify patients into immune subgroups, which were associated with differential melanoma specific survival and potential predictive biomarkers. However, acquiring transcriptome data is a time-consuming and costly process. Moreover, it is not routinely used in the current clinical workflow. Here, we attempt to overcome this by developing deep learning models to classify gigapixel haematoxylin and eosin (H&E) stained pathology slides, which are well established in clinical workflows, into these immune subgroups. We systematically assess six different multiple instance learning (MIL) frameworks, using five different image resolutions and three different feature extraction methods. We show that pathology-specific self-supervised models using 10x resolution patches generate superior representations for the classification of immune subtypes. In addition, in a primary melanoma dataset, we achieve a mean area under the receiver operating characteristic curve (AUC) of 0.80 for classifying histopathology images into 'high' or 'low immune' subgroups and a mean AUC of 0.82 in an independent TCGA melanoma dataset. Furthermore, we show that these models are able to stratify patients into 'high' and 'low immune' subgroups with significantly different melanoma specific survival outcomes (log rank test, P< 0.005). We anticipate that MIL methods will allow us to find new biomarkers of high importance, act as a tool for clinicians to infer the immune landscape of tumours and stratify patients, without needing to carry out additional expensive genetic tests.

The role of artificial intelligence and convolutional neural networks in the management of melanoma: a clinical, pathological, and radiological perspective

Clinical dermatoscopy and pathological slide assessment are essential in the diagnosis and management of patients with cutaneous melanoma. For those presenting with stage IIC disease and beyond, radiological investigations are often considered. The dermatoscopic, whole slide and radiological images used during clinical care are often stored digitally, enabling artificial intelligence (AI) and convolutional neural networks (CNN) to learn, analyse and contribute to the clinical decision-making. A keyword search of the Medline database was performed to assess the progression, capabilities and limitations of AI and CNN and its use in diagnosis and management of cutaneous melanoma. Full-text articles were reviewed if they related to dermatoscopy, pathological slide assessment or radiology. Through analysis of 95 studies, we demonstrate that diagnostic accuracy of AI/CNN can be superior (or at least equal) to clinicians. However, variability in image acquisition, pre-processing, segmentation, and feature extraction remains challenging. With current technological abilities, AI/CNN and clinicians synergistically working together are better than one another in all subspecialty domains relating to cutaneous melanoma. AI has the potential to enhance the diagnostic capabilities of junior dermatology trainees, primary care skin cancer clinicians and general practitioners. For experienced clinicians, AI provides a cost-efficient second opinion. From a pathological and radiological perspective, CNN has the potential to improve workflow efficiency, allowing clinicians to achieve more in a finite amount of time. Until the challenges of AI/CNN are reliably met, however, they can only remain an adjunct to clinical decision-making.

Nomograms based on the lymphocyte-albumin-neutrophil ratio (LANR) for predicting the prognosis of nasopharyngeal carcinoma patients after definitive radiotherapy

Much evidence has accumulated to show that inflammation and nutritional status are associated with the prognosis of patients with various cancers. The present study was designed to explore the prognostic role of the LANR in NPC patients receiving definitive radiotherapy and to construct a nomogram for predicting patient survival. This study retrospectively reviewed 805 NPC patients (604 in the training cohort and 201 in the validation cohort) who received definitive radiotherapy between January 2013 and December 2019. The clinical data and pretreatment laboratory test data, including lymphocyte count, neutrophil count, and serum ALB concentration, were collected for all patients. The LANR was calculated as the albumin × lymphocyte/neutrophil ratio. Patients in the training cohort and validation cohort were categorized into high-LANR and low-LANR groups according to the corresponding cutoff values. The independent prognostic factors for overall survival (OS), progression-free survival (PFS), relapse-free survival (RFS), and metastasis-free survival (MFS) were evaluated by univariate and multivariate Cox regression analyses, and a nomogram was subsequently constructed. The performance of the nomogram was evaluated by the concordance index (C-index) and calibration curve. A low LANR (< 14.3) was independently associated with worse OS, PFS and MFS in NPC patients. A prognostic prediction nomogram was established based on T stage, N stage, Eastern Cooperative Oncology Group (ECOG) score, treatment modality, and LANR and was validated. The C-indices of the nomograms for OS and PFS in the training cohort were 0.729 and 0.72, respectively. The C-indices of the nomograms for OS and PFS in the validation cohort were 0.694 and 0.695, respectively. The calibration curve revealed good consistency between the actual survival and the nomogram prediction. Patients with NPC with low pretreatment LANR had a poor prognosis. The nomogram established on the basis of the LANR was efficient and clinically useful for predicting survival in NPC patients who underwent definitive radiotherapy.

Prognostic and predictive biomarkers in melanoma

Biomarkers help to inform the clinical management of patients with melanoma. For patients with clinically localised primary melanoma, biomarkers can help to predict post-surgical outcome (including via the use of risk prediction tools), better select patients for sentinel lymph node biopsy, and tailor catch-all follow-up protocols to the individual. Systemic drug treatments, including immune checkpoint inhibitor (ICI) therapies and BRAF-targeted therapies, have radically improved the prognosis of metastatic (stage III and IV) cutaneous melanoma patients, and also shown benefit in the earlier setting of stage IIB/C primary melanoma. Unfortunately, a response is far from guaranteed. Here, we review clinically relevant, established, and emerging, prognostic, and predictive pathological biomarkers that refine clinical decision-making in primary and metastatic melanoma patients. Gene expression profile assays and nomograms are emerging tools for prognostication and sentinel lymph node risk prediction in primary melanoma patients. Biomarkers incorporated into clinical practice guidelines include BRAF V600 mutations for the use of targeted therapies in metastatic cutaneous melanoma, and the HLA-A∗02:01 allele for the use of a bispecific fusion protein in metastatic uveal melanoma. Several predictive biomarkers have been proposed for ICI therapies but have not been incorporated into Australian clinical practice guidelines. Further research, validation, and assessment of clinical utility is required before more prognostic and predictive biomarkers are fluidly integrated into routine care.