Digital pathology and artificial intelligence in translational medicine and clinical practice.

Traditional pathology approaches have played an integral role in the delivery of diagnosis, semi-quantitative or qualitative assessment of protein expression, and classification of disease. Technological advances and the increased focus on precision medicine have recently paved the way for the development of digital pathology-based approaches for quantitative pathologic assessments, namely whole slide imaging and artificial intelligence (AI)-based solutions, allowing us to explore and extract information beyond human visual perception. Within the field of immuno-oncology, the application of such methodologies in drug development and translational research have created invaluable opportunities for deciphering complex pathophysiology and the discovery of novel biomarkers and drug targets. With an increasing number of treatment options available for any given disease, practitioners face the growing challenge of selecting the most appropriate treatment for each patient. The ever-increasing utilization of AI-based approaches substantially expands our understanding of the tumor microenvironment, with digital approaches to patient stratification and selection for diagnostic assays supporting the identification of the optimal treatment regimen based on patient profiles. This review provides an overview of the opportunities and limitations around implementing AI-based methods in biomarker discovery and patient selection and discusses how advances in digital pathology and AI should be considered in the current landscape of translational medicine, touching on challenges this technology may face if adopted in clinical settings. The traditional role of pathologists in delivering accurate diagnoses or assessing biomarkers for companion diagnostics may be enhanced in precision, reproducibility, and scale by AI-powered analysis tools.

Association of artificial intelligence-powered and manual quantification of programmed death-ligand 1 (PD-L1) expression with outcomes in patients treated with nivolumab ± ipilimumab.

Assessment of programmed death ligand 1 (PD-L1) expression by immunohistochemistry (IHC) has emerged as an important predictive biomarker across multiple tumor types. However, manual quantitation of PD-L1 positivity can be difficult and leads to substantial inter-observer variability. Although the development of artificial intelligence (AI) algorithms may mitigate some of the challenges associated with manual assessment and improve the accuracy of PD-L1 expression scoring, use of AI-based approaches to oncology biomarker scoring and drug development has been sparse, primarily due to the lack of large-scale clinical validation studies across multiple cohorts and tumor types. We developed AI-powered algorithms to evaluate PD-L1 expression on tumor cells by IHC and compared it with manual IHC scoring in urothelial carcinoma, non-small cell lung cancer, melanoma, and squamous cell carcinoma of the head and neck (prospectively determined during the phase II and III CheckMate clinical trials). 1,746 slides were retrospectively analyzed, the largest investigation of digital pathology algorithms on clinical trial datasets performed to date. AI-powered quantification of PD-L1 expression on tumor cells identified more PD-L1-positive samples compared with manual scoring at cutoffs of ≥1% and ≥5% in most tumor types. Additionally, similar improvements in response and survival were observed in patients identified as PD-L1-positive compared with PD-L1-negative using both AI-powered and manual methods, while improved associations with survival were observed in patients with certain tumor types identified as PD-L1-positive using AI-powered scoring only. Our study demonstrates the potential for implementation of digital pathology-based methods in future clinical practice to identify more patients who would benefit from treatment with immuno-oncology therapy compared with current guidelines using manual assessment.

A Machine Learning Approach to Liver Histological Evaluation Predicts Clinically Significant Portal Hypertension in NASH Cirrhosis.

BACKGROUND AND AIMS: The hepatic venous pressure gradient (HVPG) is the standard for estimating portal pressure but requires expertise for interpretation. We hypothesized that HVPG could be extrapolated from liver histology using a machine learning (ML) algorithm. APPROACH AND RESULTS: Patients with NASH with compensated cirrhosis from a phase 2b trial were included. HVPG and biopsies from baseline and weeks 48 and 96 were reviewed centrally, and biopsies evaluated with a convolutional neural network (PathAI, Boston, MA). Using trichrome-stained biopsies in the training set (n = 130), an ML model was developed to recognize fibrosis patterns associated with HVPG, and the resultant ML HVPG score was validated in a held-out test set (n = 88). Associations between the ML HVPG score with measured HVPG and liver-related events, and performance of the ML HVPG score for clinically significant portal hypertension (CSPH) (HVPG ≥ 10 mm Hg), were determined. The ML-HVPG score was more strongly correlated with HVPG than hepatic collagen by morphometry (ρ = 0.47 vs. ρ = 0.28; P < 0.001). The ML HVPG score differentiated patients with normal (0-5 mm Hg) and elevated (5.5-9.5 mm Hg) HVPG and CSPH (median: 1.51 vs. 1.93 vs. 2.60; all P < 0.05). The areas under receiver operating characteristic curve (AUROCs) (95% CI) of the ML-HVPG score for CSPH were 0.85 (0.80, 0.90) and 0.76 (0.68, 0.85) in the training and test sets, respectively. Discrimination of the ML-HVPG score for CSPH improved with the addition of a ML parameter for nodularity, Enhanced Liver Fibrosis, platelets, aspartate aminotransferase (AST), and bilirubin (AUROC in test set: 0.85; 95% CI: 0.78, 0.92). Although baseline ML-HVPG score was not prognostic, changes were predictive of clinical events (HR: 2.13; 95% CI: 1.26, 3.59) and associated with hemodynamic response and fibrosis improvement. CONCLUSIONS: An ML model based on trichrome-stained liver biopsy slides can predict CSPH in patients with NASH with cirrhosis.

A Machine Learning Approach Enables Quantitative Measurement of Liver Histology and Disease Monitoring in NASH.

BACKGROUND AND AIMS: Manual histological assessment is currently the accepted standard for diagnosing and monitoring disease progression in NASH, but is limited by variability in interpretation and insensitivity to change. Thus, there is a critical need for improved tools to assess liver pathology in order to risk stratify NASH patients and monitor treatment response. APPROACH AND RESULTS: Here, we describe a machine learning (ML)-based approach to liver histology assessment, which accurately characterizes disease severity and heterogeneity, and sensitively quantifies treatment response in NASH. We use samples from three randomized controlled trials to build and then validate deep convolutional neural networks to measure key histological features in NASH, including steatosis, inflammation, hepatocellular ballooning, and fibrosis. The ML-based predictions showed strong correlations with expert pathologists and were prognostic of progression to cirrhosis and liver-related clinical events. We developed a heterogeneity-sensitive metric of fibrosis response, the Deep Learning Treatment Assessment Liver Fibrosis score, which measured antifibrotic treatment effects that went undetected by manual pathological staging and was concordant with histological disease progression. CONCLUSIONS: Our ML method has shown reproducibility and sensitivity and was prognostic for disease progression, demonstrating the power of ML to advance our understanding of disease heterogeneity in NASH, risk stratify affected patients, and facilitate the development of therapies.

Telling the difference and the telling differences between hospital in the home and outpatient parenteral antibiotic therapy.

Doctors, authors, funders and hospital managers should take care to distinguish the important differences between hospital in the home (HIH) and outpatient parenteral antimicrobial therapy (OPAT) services. HIH is an inpatient service delivered at home usually by (or on behalf of) hospitals, which aims to substitute for a traditional inpatient stay. It does so by delivering a wide range of hospital treatments to patients at home, or residential aged care, using hospital medical and nursing staff, delivery technologies and venous access, pharmacy, radiology and pathology, and a structured system of on call and governance. OPAT is an outpatient service, usually run through infectious diseases physicians' offices or departments. Most care is delivered in infusion centres and requires patients to travel for their care. Generally, there is no after-hours support. HIH has supplanted the role of OPAT due to improved governance and a wider clinical and severity scope. HIH is accessible from hospital emergency departments or directly from residential aged care facilities. Inpatient capacity has been expanded during the COVID-19 pandemic. There is evidence that both HIH and OPAT can successfully treat their selected patient groups. There are no head-to-head studies, but in observational comparisons there might be more adverse drug events in OPAT. OPAT places a greater onus of care, supervision and travel needs on the patient and family. Where HIH is not available, OPAT may remain an alternative for some patients. However, HIH seeks to redefine the delivery of inpatient care away from the location of care.

Home ward bound: features of hospital in the home use by major Australian hospitals, 2011-2017.

OBJECTIVE: To describe uptake of hospital in the home (HIH) by major Australian hospitals and the characteristics of patients and their HIH admissions; to assess change in HIH admission numbers relative to total hospital activity. DESIGN: Descriptive, retrospective study of HIH activity, analysing previously collected census data for all multi-day hospital inpatient admissions to included hospitals during the period 1 January 2011 - 31 December 2017. SETTING, PARTICIPANTS: Nineteen principal referrer hospital members of the Health Roundtable in Australia. MAIN OUTCOME MEASURES: HIH admissions by diagnosis-related group (DRG); patient and admission characteristics. RESULTS: 80 167 of 2 185 421 admissions to the 19 hospitals included HIH care, or 3.7% (95% CI, 3.6-3.7%) of all admissions. Median length of stay for admissions including HIH (7.3 days; IQR, 3.1-14 days) was longer than that for those that did not (2.7 days; IQR, 1.6-5.1 days). For HIH admissions, the proportion of men was higher (54.4% v 45.9%), the proportion of patients who died in hospital was lower (0.3% v 1.4%), and re-admission within 28 days was less frequent (2.3% v 3.6%). The 50 DRGs with greatest HIH activity encompassed 65 811 HIH admissions (82.1%), or 8.4% (95% CI, 8.4-8.5%) of all admissions in these DRGs. HIH admission numbers grew more rapidly than non-HIH admissions, but the difference was not statistically significant. CONCLUSIONS: HIH care is most frequently provided to patients requiring hospital treatment related to infections, venous thromboembolism, or post-surgical care. Its use could be expanded in clinical areas where it is currently used, and extended to others where it is not. HIH activity is growing. It should be systematically monitored and reported to allow better overview of its use and outcomes.

Highly multiplexed single-cell analysis of formalin-fixed, paraffin-embedded cancer tissue.

Limitations on the number of unique protein and DNA molecules that can be characterized microscopically in a single tissue specimen impede advances in understanding the biological basis of health and disease. Here we present a multiplexed fluorescence microscopy method (MxIF) for quantitative, single-cell, and subcellular characterization of multiple analytes in formalin-fixed paraffin-embedded tissue. Chemical inactivation of fluorescent dyes after each image acquisition round allows reuse of common dyes in iterative staining and imaging cycles. The mild inactivation chemistry is compatible with total and phosphoprotein detection, as well as DNA FISH. Accurate computational registration of sequential images is achieved by aligning nuclear counterstain-derived fiducial points. Individual cells, plasma membrane, cytoplasm, nucleus, tumor, and stromal regions are segmented to achieve cellular and subcellular quantification of multiplexed targets. In a comparison of pathologist scoring of diaminobenzidine staining of serial sections and automated MxIF scoring of a single section, human epidermal growth factor receptor 2, estrogen receptor, p53, and androgen receptor staining by diaminobenzidine and MxIF methods yielded similar results. Single-cell staining patterns of 61 protein antigens by MxIF in 747 colorectal cancer subjects reveals extensive tumor heterogeneity, and cluster analysis of divergent signaling through ERK1/2, S6 kinase 1, and 4E binding protein 1 provides insights into the spatial organization of mechanistic target of rapamycin and MAPK signal transduction. Our results suggest MxIF should be broadly applicable to problems in the fields of basic biological research, drug discovery and development, and clinical diagnostics.

The treatment of nursing home-acquired pneumonia using a medically intensive Hospital in the Home service.

OBJECTIVES: To compare the outcomes for patients with nursing home-acquired pneumonia (NHAP) treated completely in a Hospital in the Home (HITH) setting with those of patients treated in a traditional hospital ward. DESIGN: Case-control study. SETTING AND PARTICIPANTS: All patients admitted by the Royal Melbourne Hospital for treatment of NHAP from 1 July 2013 to 31 January 2014. INTERVENTION: Admission to the Royal Melbourne Hospital HITH Unit within 48 hours of presentation. MAIN OUTCOME MEASURES: Length of stay, in-hospital and 30-day mortality, hospital readmissions (30-day), complications and unplanned returns to hospital. RESULTS: Sixty HITH patients and 54 hospital (control) patients were identified. Thirty-two patients (53%) were admitted directly to HITH without any hospital or emergency stay, 25 (42%) were referred directly from the emergency department. HITH patients were more likely to be male, older and dehydrated, and less likely to have an advanced care directive or to have had non-invasive ventilation. There were no significant differences in CURB-65 or CORB scores between the two patient groups; similar proportions were given intravenous fluids or supplemental oxygen. There were no adjusted differences in median length of stay between HITH and control patients (-1.00 days; 95% CI, -2.72 to 0.72; P = 0.252) or in overall mortality at 30 days (HITH v control patients: adjusted odds ratio [aOR], 1.97; 95% CI, 0.67-5.73). Inpatient mortality was lower for HITH patients (aOR, 0.19; 95% CI, 0.05-0.75) but unadjusted postdischarge 30-day mortality was higher (OR, 13.25; 95% CI 1.67-105.75). There were no differences between the two groups with regard to complications (falls and pressure wounds) and 30-day readmission rates (aOR, 1.59; 95% CI, 0.30-8.53). CONCLUSIONS: This study suggests that HITH may be an effective and safe alternative to hospital treatment for residents of aged care facilities presenting with NHAP.

Evaluation of a mobile X-ray service for elderly residents of residential aged care facilities.

OBJECTIVE: The Royal Melbourne Hospital established a mobile X-ray service (MXS) in 2013. The goal of the MXS is to address the radiology needs of frail, elderly or demented residents of residential aged care facilities (RACFs) who would otherwise require transportation to attend for X-ray. The present study describes the activity of the MXS, and the impact of the MXS on emergency department (ED) attendances by residents of RACFs. METHODS: The study is a descriptive study and uses a before-and-after cohort approach. Activity for the first year of operation was collected and described. At the end of the first year of operation, the top 30 RACF users of the MXS were identified. The hospital Department of Radiology database was examined to find all plain X-rays performed on any patient presenting from the same 30 RACFs for the 1 year before commencement of the MXS (1 July 2012-30 June 2013) and for the 1 year period after the commencement of the MXS (1 July 2013-30 June 2014). Attendances were compared. RESULTS: The MXS delivered 1532 service attendances to 109 different RACFs. The mean age of patients receiving MXS services was 86 years (range 16-107 years). In all, 1124 services (73.4%) were delivered to patients in high-care RACFs. Most patients (n = 634; 41.4%) were bed or wheelchair bound, followed by those who required assistance to ambulate (n = 457; 29.8%). The most common X-ray examinations performed were chest, hip and pelvis, spine and abdomen. There were 919 service attendances to the top 30 RACFs using the MXS (60.0% of all attendances). There was an 11.5% reduction in ED presentations requiring plain X-ray in the year following the commencement of the MXS (95% confidence interval 0.62-3.98; P = 0.019). CONCLUSION: The present study suggests a reduction in hospital ED attendances for high users of the MXS. This has benefits for hospitals, patients and nursing homes. It also allows the extension of other programs designed to treat patients in their RACFs. Special rebates for home-based radiology service provision should be considered.

Hospital at home worldwide: Program and clinician characteristics from the World Hospital at Home Congress survey.

BACKGROUND: Hospital at home (HaH) delivers hospital-level care to acutely ill patients at home as a substitute for brick-and-mortar hospital care. The clinician and program characteristics of HaH programs worldwide are relatively unknown. We sought to describe the world's HaH clinicians and their programs' characteristics. METHODS: We analyzed a survey administered to all attendees of the 2023 World Hospital at Home Congress. Clinician characteristics included age, years worked in HaH, profession, burnout, and experience. Program characteristics included location, daily census, types of care delivery, and clinical capabilities. RESULTS: Of 670 attendees, about 305 were clinicians and 129 responded (42% response rate for clinicians). The majority of clinicians were 30-49 years old (65.1%), new to the field (70.5% worked less than 10 years), and part-time (18% dedicated >74% effort to HaH). Clinicians reported overall satisfaction with their job and low burnout. About half of programs were in Europe (52.1%), newly operational (44.7% less than 5 years), mostly operated in urban environments (87.2%), and mostly had a daily census of less than 25 patients (62.8%). Most programs operated 7-days per week (88.3%), performed intermittent or continuous remote monitoring (81.4%), used video communication (63.8%), and had some advanced capabilities such as in-home imaging (47.9%) and advanced procedures (23.4%). Visit frequencies to the patient's home were variable: most programs had physicians visit the home, nearly all had nurses visit the home, and fewer performed virtual visits. CONCLUSIONS: HaH clinicians and programs have significant similarities but also a fair number of divergent practices, much like brick-and-mortar hospital care. Further standardization of the care model will help to unify the field across the globe.

AI powered quantification of nuclear morphology in cancers enables prediction of genome instability and prognosis.

While alterations in nucleus size, shape, and color are ubiquitous in cancer, comprehensive quantification of nuclear morphology across a whole-slide histologic image remains a challenge. Here, we describe the development of a pan-tissue, deep learning-based digital pathology pipeline for exhaustive nucleus detection, segmentation, and classification and the utility of this pipeline for nuclear morphologic biomarker discovery. Manually-collected nucleus annotations were used to train an object detection and segmentation model for identifying nuclei, which was deployed to segment nuclei in H&E-stained slides from the BRCA, LUAD, and PRAD TCGA cohorts. Interpretable features describing the shape, size, color, and texture of each nucleus were extracted from segmented nuclei and compared to measurements of genomic instability, gene expression, and prognosis. The nuclear segmentation and classification model trained herein performed comparably to previously reported models. Features extracted from the model revealed differences sufficient to distinguish between BRCA, LUAD, and PRAD. Furthermore, cancer cell nuclear area was associated with increased aneuploidy score and homologous recombination deficiency. In BRCA, increased fibroblast nuclear area was indicative of poor progression-free and overall survival and was associated with gene expression signatures related to extracellular matrix remodeling and anti-tumor immunity. Thus, we developed a powerful pan-tissue approach for nucleus segmentation and featurization, enabling the construction of predictive models and the identification of features linking nuclear morphology with clinically-relevant prognostic biomarkers across multiple cancer types.

AI-based automation of enrollment criteria and endpoint assessment in clinical trials in liver diseases.

Clinical trials in metabolic dysfunction-associated steatohepatitis (MASH, formerly known as nonalcoholic steatohepatitis) require histologic scoring for assessment of inclusion criteria and endpoints. However, variability in interpretation has impacted clinical trial outcomes. We developed an artificial intelligence-based measurement (AIM) tool for scoring MASH histology (AIM-MASH). AIM-MASH predictions for MASH Clinical Research Network necroinflammation grades and fibrosis stages were reproducible (κ = 1) and aligned with expert pathologist consensus scores (κ = 0.62-0.74). The AIM-MASH versus consensus agreements were comparable to average pathologists for MASH Clinical Research Network scores (82% versus 81%) and fibrosis (97% versus 96%). Continuous scores produced by AIM-MASH for key histological features of MASH correlated with mean pathologist scores and noninvasive biomarkers and strongly predicted progression-free survival in patients with stage 3 (P < 0.0001) and stage 4 (P = 0.03) fibrosis. In a retrospective analysis of the ATLAS trial (NCT03449446), responders receiving study treatment showed a greater continuous change in fibrosis compared with placebo (P = 0.02). Overall, these results suggest that AIM-MASH may assist pathologists in histologic review of MASH clinical trials, reducing inter-rater variability on trial outcomes and offering a more sensitive and reproducible measure of patient responses.

Skin surface temperature: a possible new outcome measure for skin and soft tissue infection.

BACKGROUND: This study describes the novel use of skin surface temperature to measure the severity and the response to treatment of skin and soft tissue infection (SSTI). METHODS: Patients admitted with SSTI for intravenous antibiotic therapy. Skin temperature was measured daily at the point of maximum heat on the SSTI affected limb and the non-affected limb using a non-contact laser thermometer. FINDINGS: Sixty-three patients were included. Mean length of stay was 4.95 days. The difference between affected and unaffected limb was 3.5° C (95% CI 3.0-3.9) at day one and 2.1° C (95% CI 1.7-2.6) on the last day, a difference of 1.4° C (95% CI 0.7-1.9). Between day one and the last day, there was a significant reduction in affected limb temperature (mean reduction of 2.4° C, 95% CI 1.9-3.0 p<0.001). INTERPRETATION: Skin surface temperature may hold a useful role in the management of SSTI.

Integration of deep learning-based histopathology and transcriptomics reveals key genes associated with fibrogenesis in patients with advanced NASH.

Nonalcoholic steatohepatitis (NASH) is the most common chronic liver disease globally and a leading cause for liver transplantation in the US. Its pathogenesis remains imprecisely defined. We combined two high-resolution modalities to tissue samples from NASH clinical trials, machine learning (ML)-based quantification of histological features and transcriptomics, to identify genes that are associated with disease progression and clinical events. A histopathology-driven 5-gene expression signature predicted disease progression and clinical events in patients with NASH with F3 (pre-cirrhotic) and F4 (cirrhotic) fibrosis. Notably, the Notch signaling pathway and genes implicated in liver-related diseases were enriched in this expression signature. In a validation cohort where pharmacologic intervention improved disease histology, multiple Notch signaling components were suppressed.

AI-based histologic scoring enables automated and reproducible assessment of enrollment criteria and endpoints in NASH clinical trials.

Clinical trials in nonalcoholic steatohepatitis (NASH) require histologic scoring for assessment of inclusion criteria and endpoints. However, guidelines for scoring key features have led to variability in interpretation, impacting clinical trial outcomes. We developed an artificial intelligence (AI)-based measurement (AIM) tool for scoring NASH histology (AIM-NASH). AIM-NASH predictions for NASH Clinical Research Network (CRN) grades of necroinflammation and stages of fibrosis aligned with expert consensus scores and were reproducible. Continuous scores produced by AIM-NASH for key histological features of NASH correlated with mean pathologist scores and with noninvasive biomarkers and strongly predicted patient outcomes. In a retrospective analysis of the ATLAS trial, previously unmet pathological endpoints were met when scored by the AIM-NASH algorithm alone. Overall, these results suggest that AIM-NASH may assist pathologists in histologic review of NASH clinical trials, reducing inter-rater variability on trial outcomes and offering a more sensitive and reproducible measure of patient therapeutic response.

A research agenda for hospital at home.

BACKGROUND: Hospital at home (HaH) provides hospital-level care at home as a substitute for traditional hospital care. Interest in HaH is increasing markedly. While multiple studies of HaH have demonstrated that HaH provides safe, high-quality, cost-effective care, there remain many unanswered research questions. The objective of this study is to develop a research agenda to guide future HaH-related research. METHODS: Survey of attendees of first World HaH Congress 2019 for input on research for the future HaH development. Selection and ranking of important topic areas for future HaH-related research. Development of research domains and research questions and issues using grounded theory approach, supplemented by focused literature reviews. RESULTS: 240 conference attendees responded to the survey (response rate, 55.3%). The majority were from Europe (64%) and North America (11%) and were HaH program leaders (29%), HaH physicians (27%), and researchers (13%). Nine research domains for future HaH research were identified: 1) definition of the HaH model of care; 2) the HaH clinical model; 3) measurement and outcomes of HaH; 4) patient and caregiver experience with HaH; 5) education and training of HaH clinicians; 6) technology and telehealth for HaH; 7) regulatory and payment issues in HaH; 8) implementation and scaling of HaH; and 9) ethical issues in HaH. Key research issues and questions were identified for each domain. CONCLUSIONS: While highly evidence-based, unanswered research questions regarding HaH remain, focusing research efforts on the domains identified in this study will serve to improve HaH for all key HaH stakeholders.