Prediction of immunochemotherapy response for diffuse large B-cell lymphoma using artificial intelligence digital pathology

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Validation of the 2022 European LeukemiaNet risk stratification for acute myeloid leukemia

This study aimed to validate the 2022 European LeukemiaNet (ELN) risk stratification for acute myeloid leukemia (AML). A total of 624 newly diagnosed AML patients from 1998 to 2014 were included in the analysis. Genetic profiling was conducted using targeted deep sequencing of 45 genes based on recurrent driver mutations. In total, 134 (21.5%) patients had their risk classification reassessed according to the 2022 ELN risk stratification. Among those initially classified as having a favorable risk in 2017 (n = 218), 31 and 3 patients were reclassified as having intermediate risk or adverse risk, respectively. Among the three subgroups, the 2022 ELN favorable-risk group showed significantly longer survival outcomes than the other groups. Within the 2017 ELN intermediate-risk group (n = 298), 21 and 46 patients were reclassified as having favorable risk or adverse risk, respectively, and each group showed significant stratifications in survival outcomes. Some patients initially classified as having adverse risk in 2017 were reclassified into the intermediate-risk group (33 of 108 patients), but no prognostic improvements were observed in this group. A multivariable analysis identified the 2022 ELN risk stratification, age, and receiving allogeneic hematopoietic cell transplantation as significant prognostic factors for survival. The 2022 ELN risk stratification enables more precise decisions for proceeding with allogeneic hematopoietic cell transplantation for AML patients.

Ensemble Deep Learning Model to Predict Lymphovascular Invasion in Gastric Cancer

Lymphovascular invasion (LVI) is one of the most important prognostic factors in gastric cancer as it indicates a higher likelihood of lymph node metastasis and poorer overall outcome for the patient. Despite its importance, the detection of LVI(+) in histopathology specimens of gastric cancer can be a challenging task for pathologists as invasion can be subtle and difficult to discern. Herein, we propose a deep learning-based LVI(+) detection method using H&E-stained whole-slide images. The ConViT model showed the best performance in terms of both AUROC and AURPC among the classification models (AUROC: 0.9796; AUPRC: 0.9648). The AUROC and AUPRC of YOLOX computed based on the augmented patch-level confidence score were slightly lower (AUROC: -0.0094; AUPRC: -0.0225) than those of the ConViT classification model. With weighted averaging of the patch-level confidence scores, the ensemble model exhibited the best AUROC, AUPRC, and F1 scores of 0.9880, 0.9769, and 0.9280, respectively. The proposed model is expected to contribute to precision medicine by potentially saving examination-related time and labor and reducing disagreements among pathologists.

CD5 Expression Dynamically Changes During the Differentiation of Human CD8+ T Cells Predicting Clinical Response to Immunotherapy

Defining the molecular dynamics associated with T cell differentiation enhances our understanding of T cell biology and opens up new possibilities for clinical implications. In this study, we investigated the dynamics of CD5 expression in CD8+ T cell differentiation and explored its potential clinical uses. Using PBMCs from 29 healthy donors, we observed a stepwise decrease in CD5 expression as CD8+ T cells progressed through the differentiation stages. Interestingly, we found that CD5 expression was initially upregulated in response to T cell receptor stimulation, but diminished as the cells underwent proliferation, potentially explaining the differentiation-associated CD5 downregulation. Based on the proliferation-dependent downregulation of CD5, we hypothesized that relative CD5 expression could serve as a marker to distinguish the heterogeneous CD8+ T cell population based on their proliferation history. In support of this, we demonstrated that effector memory CD8+ T cells with higher CD5 expression exhibited phenotypic and functional characteristics resembling less differentiated cells compared to those with lower CD5 expression. Furthermore, in the retrospective analysis of PBMCs from 30 non-small cell lung cancer patients, we found that patients with higher CD5 expression in effector memory T cells displayed CD8+ T cells with a phenotype closer to the less differentiated cells, leading to favorable clinical outcomes in response to immune checkpoint inhibitor (ICI) therapy. These findings highlight the dynamics of CD5 expression as an indicator of CD8+ T cell differentiation status, and have implications for the development of predictive biomarker for ICI therapy.

Effect of preoperative immunonutrition on fecal microbiota in colon cancer patients: a secondary analysis of a randomized controlled trial

BACKGROUND/OBJECTIVES

This study aimed to evaluate the effect of preoperative immunonutrition on the composition of fecal microbiota following a colon cancer surgery.

MATERIALS/METHODS

This study was a secondary analysis of a randomized controlled trial assessing the impact of preoperative immunonutrition on the postoperative outcomes of colon cancer surgery. Patients with primary colon cancer were enrolled and randomly assigned to receive additional preoperative immunonutrition or a normal diet alone. Oral nutritional supplementation (400 mL/day) with arginine and ω-3 fatty acids were administered to patients in the immunonutrition group for 7 days prior to surgery. Thirty-two fecal samples were collected from 16 patients in each group, and the composition of fecal microbiota was compared between the 2 groups.

RESULTS

At the phylum level, no significant difference was observed in the composition of microbiota between the 2 groups (Firmicutes, 69.1% vs. 67.5%, P = 0.624; Bacteroidetes, 19.3% vs. 18.1%, P = 0.663; Actinobacteria, 6.7% vs. 10.6%, P = 0.080). The Firmicutes/Bacteroidetes ratio (4.43 ± 2.32 vs. 4.55 ± 2.51, P = 0.897) was also similar between the 2 groups. At the genus level, the proportions of beneficial bacteria such as Faecalibacterium spp. (8.1% vs. 6.4%, P = 0.328) and Prevotella spp. (6.9% vs. 4.8%, P = 0.331) were higher, while that of Clostridium spp. was lower (0.5% vs. 1.2%, P = 0.121) in the immunonutrition group, but the difference was not significant.

CONCLUSIONS

Immunonutrition showed no significant association with the composition of fecal microbiota. The relationship between immunonutrition and the fecal microbiota should be investigated further in large-scale studies.

Fosfomycin for Antibiotic Prophylaxis in Men Undergoing a Transrectal Prostate Biopsy: A Systematic Review and Meta-Analysis

Background and Objectives: To assess the effects of fosfomycin compared with other antibiotics as a prophylaxis for urinary tract infections (UTIs) in men undergoing transrectal prostate biopsies. Materials and Methods: We searched multiple databases and trial registries without publication language or status restrictions until 4 January 2022. Parallel-group randomized controlled trials (RCTs) and non-randomized studies (NRS) were included. The primary outcomes were febrile UTI, afebrile UTI, and overall UTI. We used GRADE guidance to rate the certainty of evidence of RCTs and NRSs. The protocol was registered with PROSPERO (CRD42022302743). Results: We found data on five comparisons; however, this abstract focuses on the primary outcomes of the two most clinically relevant comparisons. Regarding fosfomycin versus fluoroquinolone, five RCTs and four NRSs with a one-month follow-up were included. Based on the RCT evidence, fosfomycin likely resulted in little to no difference in febrile UTIs compared with fluoroquinolone. This difference corresponded to four fewer febrile UTIs per 1000 patients. Fosfomycin likely resulted in little to no difference in afebrile UTIs compared with fluoroquinolone. This difference corresponded to 29 fewer afebrile UTIs per 1000 patients. Fosfomycin likely resulted in little to no difference in overall UTIs compared with fluoroquinolone. This difference corresponded to 35 fewer overall UTIs per 1000 patients. Regarding fosfomycin and fluoroquinolone combined versus fluoroquinolone, two NRSs with a one- to three-month follow-up were included. Based on the NRS evidence, fosfomycin and fluoroquinolone combined may result in little to no difference in febrile UTIs compared with fluoroquinolone. This difference corresponded to 16 fewer febrile UTIs per 1000 patients. Conclusions: Compared with fluoroquinolone, fosfomycin or fosfomycin and fluoroquinolone combined may have a similar prophylactic effect on UTIs after a transrectal prostate biopsy. Given the increasing fluoroquinolone resistance and its ease to use, fosfomycin may be a good option for antibiotic prophylaxis.

Lipocalin-2 Secreted by the Liver Regulates Neuronal Cell Function Through AKT-Dependent Signaling in Hepatic Encephalopathy Mouse Model

Hepatic encephalopathy (HE) associated with liver failure is accompanied by hyperammonemia, severe inflammation, depression, anxiety, and memory deficits as well as liver injury. Recent studies have focused on the liver-brain-inflammation axis to identify a therapeutic solution for patients with HE. Lipocalin-2 is an inflammation-related glycoprotein that is secreted by various organs and is involved in cellular mechanisms including iron homeostasis, glucose metabolism, cell death, neurite outgrowth, and neurogenesis. In this study, we investigated that the roles of lipocalin-2 both in the brain cortex of mice with HE and in Neuro-2a (N2A) cells. We detected elevated levels of lipocalin-2 both in the plasma and liver in a bile duct ligation mouse model of HE. We confirmed changes in cytokine expression, such as interleukin-1β, cyclooxygenase 2 expression, and iron metabolism related to gene expression through AKT-mediated signaling both in the brain cortex of mice with HE and N2A cells. Our data showed negative effects of hepatic lipocalin-2 on cell survival, iron homeostasis, and neurite outgrowth in N2A cells. Thus, we suggest that regulation of lipocalin-2 in the brain in HE may be a critical therapeutic approach to alleviate neuropathological problems focused on the liver-brain axis.

Comparison of amikacin with fosfomycin as an add-on to ciprofloxacin for antibiotic prophylaxis in transrectal prostate biopsy: A single-center retrospective study

PURPOSE

To assess the effect of ciprofloxacin (CP) and fosfomycin compared with CP and amikacin in patients with a fluoroquinolone (FQ)-resistant rectal flora who have undergone transrectal ultrasound-guided prostate biopsy (TRUSPB).

MATERIALS AND METHODS

In total, 516 patients with FQ-resistant rectal flora based on rectal swab cultures were divided into two groups according to prophylactic antibiotics. Patients in both groups were administered CP (400 mg, intravenous [IV], twice daily) on the same day as TRUSPB and 1 day after biopsy. The amikacin group (n=260) was administered a single injection of amikacin (1 g, IV) 1 hour before TRUSPB, whereas the fosfomycin group (n=256) was administered fosfomycin (3 g, orally) the night before the procedure. The primary endpoint was the rate of infectious complications in the two groups.

RESULTS

Overall, 13 patients (2.5%) reported infectious complications: 12 patients (4.62%) in the amikacin group compared with 1 patient (0.39%) in the fosfomycin group (risk ratio, 0.09; 95% confidence interval [CI], 0.01-0.65), respectively, which was a statistically significant difference (p=0.017). This corresponds to a number needed to treat of 24 patients (95% CI, 15-65) to prevent one infectious complication. In the multivariate analysis to assess variables related to infectious complications, prophylactic antibiotics with added fosfomycin was associated with infectious complications (odds ratio, 0.060; 95% CI, 0.008-0.459).

CONCLUSIONS

In the era of FQ resistance, CP and fosfomycin may reduce the rate of infectious complications compared with CP and amikacin prophylaxis.

Association between Albumin, Total Bilirubin, and Uric Acid Serum Levels and the Risk of Cancer: A Prospective Study in a Korean Population

PURPOSE

In this prospective study, we evaluated the association between the serum levels of antioxidants uric acid (UA), albumin, and total bilirubin and the risk of cancer in a Korean population.

MATERIALS AND METHODS

A total of 15882 subjects were followed up for cancer development and cancer-related death. During the follow-up period, 1619 cancer diagnoses and 617 cancer-related deaths were recorded. Cox proportional regression was performed to calculate the hazard ratio (HR) per standard deviation (SD) increment and 95% confidence interval (CI). The model was adjusted for covariates such as the age, sex, smoking, alcohol consumption, physical activity, education level, body mass index, and family history. Sensitivity analyses using the study subjects with physiological serum levels of each indicator were also performed.

RESULTS

UA levels were positively correlated with cancer risk (HR per SD increment 1.04; 95% CI, 1.01-1.09), and albumin levels were inversely associated with the overall cancer risk (HR, 0.92; 95% CI, 0.88-0.96) and cancer-related death (HR, 0.86; 95% CI, 0.80-0.93). Total bilirubin levels were negatively correlated with the risk of cancer-related death (HR, 0.91; 95% CI, 0.83-0.99). By cancer type, UA was positively associated with prostate cancer, total bilirubin was positively associated with liver cancer, and albumin was inversely associated with lung cancer.

CONCLUSION

The findings of this study support the role of antioxidants in carcinogenesis. Future large-cohort studies are needed to confirm the predictive value of albumin, UA, and total bilirubin levels in each type of cancer.

Air and Environmental Contamination Caused by COVID-19 Patients: a Multi-Center Study

BACKGROUND

The purpose of this study was to determine the extent of air and surface contamination of severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) in four health care facilities with hospitalized coronavirus disease 2019 (COVID-19) patients.

METHODS

We investigated air and environmental contamination in the rooms of eight COVID-19 patients in four hospitals. Some patients were in negative-pressure rooms, and others were not. None had undergone aerosol-generating procedures. On days 0, 3, 5, and 7 of hospitalization, the surfaces in the rooms and anterooms were swabbed, and air samples were collected 2 m from the patient and from the anterooms.

RESULTS

All 52 air samples were negative for SARS-CoV-2 RNA. Widespread surface contamination of SARS-CoV-2 RNA was observed. In total, 89 of 320 (27%) environmental surface samples were positive for SARS-CoV-2 RNA. Surface contamination of SARS-CoV-2 RNA was common in rooms without surface disinfection and in rooms sprayed with disinfectant twice a day. However, SARS-CoV-2 RNA was not detected in a room cleaned with disinfectant wipes on a regular basis.

CONCLUSION

Our data suggest that remote (> 2 m) airborne transmission of SARS-CoV-2 from hospitalized COVID-19 patients is uncommon when aerosol-generating procedures have not been performed. Surface contamination was widespread, except in a room routinely cleaned with disinfectant wipes.

Clinical Characteristics of Korean Patients with Lung Cancer Who Have Programmed Death-Ligand 1 Expression

BACKGROUND

Programmed death-ligand 1 (PD-L1), a transmembrane protein, binds to the programmed death-1 (PD-1) receptor, and anti-PD-1 therapy enables immune responses against tumors. This study aimed to assess clinical characteristics of PD-L1 expression using immunohistochemistry among Korean patients with lung cancer.

METHODS

We retrospectively reviewed the data of patients with pathologically proven lung cancer from a single institution. PD-L1 expression determined by Tumor Proportion Score (TPS) was detected using 22C3 pharmDx (Agilent Technologies) and SP263 (Ventana Medical Systems) assays.

RESULTS

From July 2016 to July 2017, 267 patients were enrolled. The main histologic type was adenocarcinoma (69.3%). Most participants were smokers (67.4%) and had clinical stage IV disease (60.7%). In total, 116 (42%) and 58 (21%) patients had TPS ≥1% and ≥50%, respectively. The patients were significantly older in TPS ≥1% group than in TPS <1% group (64.83±9.38 years vs. 61.73±10.78 years, p=0.014), not in TPS ≥50% cutoff value (64.69 ± 9.39 vs. 62.36 ± 10.51, p= 0.178). Regarding histologic grade, higher proportions of poorly differentiated tumor were observed in the TPS ≥1% (40.8% vs. 25.8%, p=0.020) and TPS ≥50% groups (53.2% vs. 27.2%, p=0.004). Among 34 patients examined with 22C3 and SP263 assays, 27 had positive results in both assays, with a cutoff of TPS ≥1% (r=0.826; 95% confidence interval, 0.736-0.916).

CONCLUSION

PD-L1 expression, defined as TPS ≥1%, was related to older age and poorly differentiated histology. There was a similar distribution of PD-L1 expression in both 22C3 and SP263 results.

Efficacy of First-Line Targeted Therapy in Real-World Korean Patients with Metastatic Renal Cell Carcinoma: Focus on Sunitinib and Pazopanib

BACKGROUND

To evaluate survival outcomes and prognostic factors for overall survival (OS) in patients with metastatic renal cell carcinoma (mRCC) who received sunitinib (SU) and pazopanib (PZ) as first-line therapy in real-world Korean clinical practice.

METHODS

Data of 554 patients with mRCC who received SU or PZ at eight institutions between 2012 and 2016 were retrospectively reviewed. Based on the targeted therapy, the patients were divided into SU (n = 293) or PZ (n = 261) groups, and the clinicopathological variables and survival rates of the two groups were compared. A multivariable Cox proportional hazard model was used to determine the prognostic factors for OS.

RESULTS

The median follow-up was 16.4 months (interquartile range, 8.3-31.3). Patients in the PZ group were older, and no significant difference was observed in the performance status (PS) between the two groups. In the SU group, the dose reduction rate was higher and the incidence of grade 3 toxicity was more frequent. The objective response rates were comparable between the two groups (SU, 32.1% vs. PZ, 36.4%). OS did not differ significantly between the two groups (SU, 36.5 months vs. PZ, 40.2 months; log-rank, P = 0.955). Body mass index, Eastern Cooperative Oncology Group PS > 2, synchronous metastasis, poor Heng risk criteria, and liver and bone metastases were associated with a shorter OS.

CONCLUSION

Our real-world data of Korean patients with mRCC suggested that SU and PZ had similar efficacies as first-line therapy for mRCC. However, PZ was better tolerated than SU in Korean patients.

Trial Comparing a Combined Regimen of Amikacin and Ciprofloxacin to Ciprofloxacin Alone as Transrectal Prostate Biopsy Prophylaxis in the Era of High Fluoroquinolone-Resistant Rectal Flora

BACKGROUND

To investigate whether addition of amikacin to fluoroquinolone (FQ) antimicrobial prophylaxis reduces infections after transrectal ultrasound-guided prostate biopsy (TRUSPB).

METHODS

A total of 503 patients undergoing rectal swab were divided into three groups. Patients with FQ-sensitive rectal flora (group 1, n = 248) were administered ciprofloxacin before TRUSPB, and patients with FQ-resistant rectal flora were either administered ciprofloxacin (group 2, n = 97) or amikacin and ciprofloxacin (group 3, n = 158) before TRUSPB.

RESULTS

Based on the rectal swab, FQ resistance was 54.9%, and extended-spectrum β-lactamase (ESBL) positivity was 17.2%. The incidence of infectious complication in group 1 was 1.6%. Groups 2 and 3, with FQ-resistant rectal flora, tended to have increased infectious complications (5.2% and 4.4%, respectively) but the difference between those results is not statistically significant. The most common pathogens of infectious complications in patients with FQ-resistant rectal flora were FQ-resistant and ESBL-producing Escherichia coli. E. coli pathogens isolated in Group 3 were amikacin-susceptible species. The operation history and ESBL positivity of rectal flora increased the incidence of infectious complications (odds ratio [OR] = 3.68; P = 0.035 and OR = 4.02; P = 0.008, respectively). DM and antibiotics exposure were risk factors for FQ resistance (OR = 2.19; P = 0.002) and ESBL positivity of rectal flora (OR = 2.96; P = 0.005), respectively.

CONCLUSION

Addition of amikacin to ciprofloxacin prophylaxis could not reduce infectious complications in patients with FQ-resistant rectal flora. Despite the amikacin sensitivity of infectious complications, single-dose amikacin addition to ciprofloxacin prophylaxis has limitations.

The Recovery of Hypothalamic-Pituitary-Adrenal Axis Is Rapid in Subclinical Cushing Syndrome

BACKGROUND

In subclinical Cushing syndrome (SC), it is assumed that glucocorticoid production is insufficient to cause a clinically recognizable syndrome. Differences in hormonal levels or recovery time of the hypothalamic-pituitary-adrenocortical (HPA) axis after adrenalectomy between patients with overt Cushing syndrome (OC) and SC remain unknown.

METHODS

Thirty-six patients (10 with OC and 26 with SC) with adrenal Cushing syndrome who underwent adrenalectomy from 2004 to 2014 were reviewed retrospectively. Patients were treated with glucocorticoid after adrenalectomy and were reevaluated every 1 to 6 months using a rapid adrenocorticotropic hormone (ACTH) stimulation test.

RESULTS

Levels of basal 24-hour urine free cortisol (UFC), serum cortisol after an overnight dexamethasone suppression test (DST), and serum cortisol and 24-hour UFC after low-dose DST and high-dose DST were all significantly lower in patients with SC compared with OC. Basal ACTH levels showed significantly higher in patients with SC compared with OC. The probability of recovering adrenal function during follow-up differed significantly between patients with OC and SC (P=0.001), with significant correlations with the degree of preoperative cortisol excess. Patients with OC required a longer duration of glucocorticoid replacement to recover a normal ACTH stimulation test compared with patients with SC (median 17.0 months vs. 4.0 months, P<0.001).

CONCLUSION

The HPA axis recovery time after adrenalectomy in patients with SC is rapid and is dependent on the degree of cortisol excess. More precise definition of SC is necessary to achieve a better management of patients and to avoid the risk of under- or over-treatment of SC patients.