Optimizing the evaluation of gene-targeted panels for tumor mutational burden estimation

Predictive biomarkers in the era of immunotherapy for gastric cancer: current achievements and future perspectives

Gastric cancer (GC) is one of the primary contributors to cancer-related mortality on a global scale. It holds a position within the top five most prevalent malignancies both in terms of occurrence and fatality rates. Immunotherapy, as a breakthrough cancer treatment, brings new hope for GC patients. Various biomarkers, such as the expression of programmed death ligand-1 (PD-L1), the microsatellite instability (MSI) status, tumor mutational burden (TMB), and Epstein-Barr virus (EBV) infection, demonstrate potential to predict the effectiveness of immunotherapy in treating GC. Nevertheless, each biomarker has its own limitations, which leads to a significant portion of patients continue to be unresponsive to immunotherapy. With the understanding of the tumor immune microenvironment (TIME), genome sequencing technology, and recent advances in molecular biology, new molecular markers, such as POLE/POLD1mutations, circulating tumor DNA, intestinal flora, lymphocyte activation gene 3 (LAG-3), and lipid metabolism have emerged. This review aims to consolidate clinical evidence to offer a thorough comprehension of the existing and emerging biomarkers. We discuss the mechanisms, prospects of application, and limitations of each biomarker. We anticipate that this review will open avenues for fresh perspectives in the investigation of GC immunotherapy biomarkers and promote the precise choice of treatment modalities for gastric cancer patients, thereby advancing precision immuno-oncology endeavors.

Northwestern University resource and education development initiatives to advance collaborative artificial intelligence across the learning health system

Introduction

The rapid development of artificial intelligence (AI) in healthcare has exposed the unmet need for growing a multidisciplinary workforce that can collaborate effectively in the learning health systems. Maximizing the synergy among multiple teams is critical for Collaborative AI in Healthcare.

Methods

We have developed a series of data, tools, and educational resources for cultivating the next generation of multidisciplinary workforce for Collaborative AI in Healthcare. We built bulk-natural language processing pipelines to extract structured information from clinical notes and stored them in common data models. We developed multimodal AI/machine learning (ML) tools and tutorials to enrich the toolbox of the multidisciplinary workforce to analyze multimodal healthcare data. We have created a fertile ground to cross-pollinate clinicians and AI scientists and train the next generation of AI health workforce to collaborate effectively.

Results

Our work has democratized access to unstructured health information, AI/ML tools and resources for healthcare, and collaborative education resources. From 2017 to 2022, this has enabled studies in multiple clinical specialties resulting in 68 peer-reviewed publications. In 2022, our cross-discipline efforts converged and institutionalized into the Center for Collaborative AI in Healthcare.

Conclusions

Our Collaborative AI in Healthcare initiatives has created valuable educational and practical resources. They have enabled more clinicians, scientists, and hospital administrators to successfully apply AI methods in their daily research and practice, develop closer collaborations, and advanced the institution-level learning health system.

Tumor Mutation Burden as a Cornerstone in Precision Oncology Landscapes: Effect of Panel Size and Uncertainty in Cutoffs

Tumor mutation burden (TMB) has profound implications for personalized cancer therapy, particularly immunotherapy. However, the size of the panel and the cutoff values for an accurate determination of TMB are still controversial. In this study, a pan-cancer analysis was performed on 22 cancer types from The Cancer Genome Atlas. The efficiency of gene panels of different sizes and the effect of cutoff values in accurate TMB determination was assessed on a large cohort using Whole Exome Sequencing data (n = 9929 patients) as the gold standard. Gene panels of four different sizes (i.e., 0.44-2.54 Mb) were selected for comparative analyses. The heterogeneity of TMB within and between cancer types is observed to be very high, and it becomes possible to obtain the exact TMB value as the size of the panel increases. In panels with limited size, it is particularly difficult to recognize patients with low TMB. In addition, the use of a general TMB cutoff can be quite misleading. The optimal cutoff value varies between 5 and 20, depending on the TMB distribution of the different tumor types. The use of comprehensive gene panels and the optimization of TMB cutoff values for different cancer types can make TMB a robust biomarker in precision oncology. Moreover, optimization of TMB can help accelerate translational medicine research, and by extension, delivery of personalized cancer care in the future.

Informing immunotherapy with multi-omics driven machine learning

Progress in sequencing technologies and clinical experiments has revolutionized immunotherapy on solid and hematologic malignancies. However, the benefits of immunotherapy are limited to specific patient subsets, posing challenges for broader application. To improve its effectiveness, identifying biomarkers that can predict patient response is crucial. Machine learning (ML) play a pivotal role in harnessing multi-omic cancer datasets and unlocking new insights into immunotherapy. This review provides an overview of cutting-edge ML models applied in omics data for immunotherapy analysis, including immunotherapy response prediction and immunotherapy-relevant tumor microenvironment identification. We elucidate how ML leverages diverse data types to identify significant biomarkers, enhance our understanding of immunotherapy mechanisms, and optimize decision-making process. Additionally, we discuss current limitations and challenges of ML in this rapidly evolving field. Finally, we outline future directions aimed at overcoming these barriers and improving the efficiency of ML in immunotherapy research.

Tumor Mutational Burden as a Biomarker of Immunotherapy Response: An Immunogram Approach in Onco-immunology

Immune checkpoint inhibitors have revolutionized cancer treatment by allowing T cells to reactivate. Tumor mutational burden (TMB) is a biomarker that has emerged as a viable diagnostic for locating patients who would benefit from immunotherapy in particular cancer types. Greater neo-antigens mean more opportunities for T cell identification, and TMB is clinically linked to better immune checkpoint inhibitors. Tumor foreignness is a cancer immunogram, and TMB can be used as a substitute for foreignness. The role of TMB analysis as an independent predictor of immunotherapy response in the context of immune checkpoint inhibitor medications is the subject of this mini-review.

Challenges and Future Directions in the Management of Tumor Mutational Burden-High (TMB-H) Advanced Solid Malignancies

A standardized assessment of Tumor Mutational Burden (TMB) poses challenges across diverse tumor histologies, treatment modalities, and testing platforms, requiring careful consideration to ensure consistency and reproducibility. Despite clinical trials demonstrating favorable responses to immune checkpoint inhibitors (ICIs), not all patients with elevated TMB exhibit benefits, and certain tumors with a normal TMB may respond to ICIs. Therefore, a comprehensive understanding of the intricate interplay between TMB and the tumor microenvironment, as well as genomic features, is crucial to refine its predictive value. Bioinformatics advancements hold potential to improve the precision and cost-effectiveness of TMB assessments, addressing existing challenges. Similarly, integrating TMB with other biomarkers and employing comprehensive, multiomics approaches could further enhance its predictive value. Ongoing collaborative endeavors in research, standardization, and clinical validation are pivotal in harnessing the full potential of TMB as a biomarker in the clinic settings.

Restriction site associated DNA sequencing for tumour mutation burden estimation and mutation signature analysis

BACKGROUND

Genome-wide measures of genetic disruption such as tumour mutation burden (TMB) and mutation signatures are emerging as useful biomarkers to stratify patients for treatment. Clinicians commonly use cancer gene panels for tumour mutation burden estimation, and whole genome sequencing is the gold standard for mutation signature analysis. However, the accuracy and cost associated with these assays limits their utility at scale.

METHODS

WGS data from 560 breast cancer patients was used for in silico library simulations to evaluate the accuracy of an FDA approved cancer gene panel as well as restriction enzyme associated DNA sequencing (RADseq) libraries for TMB estimation and mutation signature analysis. We also transfected a mouse mammary cell line with APOBEC enzymes and sequenced resulting clones to evaluate the efficacy of RADseq in an experimental setting.

RESULTS

RADseq had improved accuracy of TMB estimation and derivation of mutation profiles when compared to the FDA approved cancer panel. Using simulated immune checkpoint blockade (ICB) trials, we show that inaccurate TMB estimation leads to a reduction in power for deriving an optimal TMB cutoff to stratify patients for immune checkpoint blockade treatment. Additionally, prioritisation of APOBEC hypermutated tumours in these trials optimises TMB cutoff determination for breast cancer. The utility of RADseq in an experimental setting was also demonstrated, based on characterisation of an APOBEC mutation signature in an APOBEC3A transfected mouse cell line.

CONCLUSION

In conclusion, our work demonstrates that RADseq has the potential to be used as a cost-effective, accurate solution for TMB estimation and mutation signature analysis by both clinicians and basic researchers.

Increased diagnostic yield from negative whole genome-slice panels using automated reanalysis

We evaluated the diagnostic yield using genome-slice panel reanalysis in the clinical setting using an automated phenotype/gene ranking system. We analyzed whole genome sequencing (WGS) data produced from clinically ordered panels built as bioinformatic slices for 16 clinically diverse, undiagnosed cases referred to the Pediatric Mendelian Genomics Research Center, an NHGRI-funded GREGoR Consortium site. Genome-wide reanalysis was performed using Moon™, a machine-learning-based tool for variant prioritization. In five out of 16 cases, we discovered a potentially clinically significant variant. In four of these cases, the variant was found in a gene not included in the original panel due to phenotypic expansion of a disorder or incomplete initial phenotyping of the patient. In the fifth case, the gene containing the variant was included in the original panel, but being a complex structural rearrangement with intronic breakpoints outside the clinically analyzed regions, it was not initially identified. Automated genome-wide reanalysis of clinical WGS data generated during targeted panels testing yielded a 25% increase in diagnostic findings and a possibly clinically relevant finding in one additional case, underscoring the added value of analyses versus those routinely performed in the clinical setting.

Mutation Burden Independently Predicts Survival in the Pan-Cancer Atlas

PURPOSE

Long-standing clinical predictors of cancer survival have included histopathologic type, stage, and grade. We hypothesized that the principal categories of tumor somatic mutations might also portend survival. We investigated this hypothesis using the Pan-Cancer Atlas, encompassing clinical, genomic, and outcome data of 10,652 patients and 32 cancer types.

METHODS

We evaluated the prognostic capability of cancer type, stage, grade and the burden of each major mutation category on overall and disease-specific survival. Mutation categories included short substitution and insertion-deletion mutations (SMs), copy number alterations (CNAs), and gene fusions.

RESULTS

SM count and CNA fraction proved to be strong independent predictors of survival (joint P = 5.3e-95) that remained highly significant when adjusted for the traditional factors. Importantly, the relationship between mutation burden and survival proved to be nonlinear (P = 9.5e-56); survival improved at both low- and high-burden extremes. In clinically predictive modeling, SM count together with CNA fraction meaningfully distinguished survival even among patients sharing a given cancer type, stage, or grade.

CONCLUSION

Burden of somatic mutation is a key index of survival of analogous clinical utility to these traditional factors.

Machine Learning for Lung Cancer Diagnosis, Treatment, and Prognosis

The recent development of imaging and sequencing technologies enables systematic advances in the clinical study of lung cancer. Meanwhile, the human mind is limited in effectively handling and fully utilizing the accumulation of such enormous amounts of data. Machine learning-based approaches play a critical role in integrating and analyzing these large and complex datasets, which have extensively characterized lung cancer through the use of different perspectives from these accrued data. In this review, we provide an overview of machine learning-based approaches that strengthen the varying aspects of lung cancer diagnosis and therapy, including early detection, auxiliary diagnosis, prognosis prediction, and immunotherapy practice. Moreover, we highlight the challenges and opportunities for future applications of machine learning in lung cancer.

Bias and inconsistency in the estimation of tumour mutation burden

BACKGROUND

Tumour mutation burden (TMB), defined as the number of somatic mutations per megabase within the sequenced region in the tumour sample, has been used as a biomarker for predicting response to immune therapy. Several studies have been conducted to assess the utility of TMB for various cancer types; however, methods to measure TMB have not been adequately evaluated. In this study, we identified two sources of bias in current methods to calculate TMB.

METHODS

We used simulated data to quantify the two sources of bias and their effect on TMB calculation, we down-sampled sequencing reads from exome sequencing datasets from TCGA to evaluate the consistency in TMB estimation across different sequencing depths. We analyzed data from ten cancer cohorts to investigate the relationship between inferred TMB and sequencing depth.

RESULTS

We found that TMB, estimated by counting the number of somatic mutations above a threshold frequency (typically 0.05), is not robust to sequencing depth. Furthermore, we show that, because only mutations with an observed frequency greater than the threshold are considered, the observed mutant allele frequency provides a biased estimate of the true frequency. This can result in substantial over-estimation of the TMB, when the cancer sample includes a large number of somatic mutations at low frequencies, and exacerbates the lack of robustness of TMB to variation in sequencing depth and tumour purity.

CONCLUSION

Our results demonstrate that care needs to be taken in the estimation of TMB to ensure that results are unbiased and consistent across studies and we suggest that accurate and robust estimation of TMB could be achieved using statistical models that estimate the full mutant allele frequency spectrum.