Retrospective analysis of the incidence of appendiceal neoplasm and malignancy in patients treated for suspected acute appendicitis.

BACKGROUND: Nonoperative management of uncomplicated appendicitis is currently being promoted as treatment option, albeit 0.7-2.5% of appendectomies performed due to suspected acute appendicitis show histologically malignant findings. The purpose of this study was to investigate the incidence of neoplasm and malignancy of the appendix in patients presenting with suspected acute appendicitis in real world setting. METHODS: This is a retrospective single-centre investigation of 457 patients undergoing appendectomy between the years 2017-2020. The patients' demographics, symptoms and diagnosis, intraoperative findings, and histopathological results were analysed. RESULTS: In 3.7% (n = 17) histological analysis revealed neoplasms or malignancies. Median age was 48 years (20-90 years), without sex predominance. Leukocytes (11.3 ± 3.7 G/l) and C-reactive protein (54.2 ± 69.0 mg/l) were elevated. Histological analysis revealed low-grade mucinous appendiceal neoplasia (n = 3), sessile serrated adenoma of the appendix (n = 3), neuroendocrine tumours (n = 7), appendiceal adenocarcinoma of intestinal type (n = 3), and goblet cell carcinoma (n = 1). Additional treatment varied between no treatment or follow-up due to early tumour stage (n = 4), follow-up care (n = 3), additional surgical treatment (n = 8), or best supportive care (n = 2). CONCLUSIONS: Preoperative diagnosis of appendiceal tumours is difficult. Nonoperative management of patients with acute, uncomplicated appendicitis potentially prevents the correct diagnosis of malignant appendiceal pathologies. Therefore, close follow-up or surgical removal of the appendix is mandatory.

International consensus on the management of metastatic gastric cancer: step by step in the foggy landscape : Bertinoro Workshop, November 2022.

BACKGROUND: Many gastric cancer patients in Western countries are diagnosed as metastatic with a median overall survival of less than twelve months using standard chemotherapy. Innovative treatments, like targeted therapy or immunotherapy, have recently proved to ameliorate prognosis, but a general agreement on managing oligometastatic disease has yet to be achieved. An international multi-disciplinary workshop was held in Bertinoro, Italy, in November 2022 to verify whether achieving a consensus on at least some topics was possible. METHODS: A two-round Delphi process was carried out, where participants were asked to answer 32 multiple-choice questions about CT, laparoscopic staging and biomarkers, systemic treatment for different localization, role and indication of palliative care. Consensus was established with at least a 67% agreement. RESULTS: The assembly agreed to define oligometastases as a "dynamic" disease which either regresses or remains stable in response to systemic treatment. In addition, the definition of oligometastases was restricted to the following sites: para-aortic nodal stations, liver, lung, and peritoneum, excluding bones. In detail, the following conditions should be considered as oligometastases: involvement of para-aortic stations, in particular 16a2 or 16b1; up to three technically resectable liver metastases; three unilateral or two bilateral lung metastases; peritoneal carcinomatosis with PCI ≤ 6. No consensus was achieved on how to classify positive cytology, which was considered as oligometastatic by 55% of participants only if converted to negative after chemotherapy. CONCLUSION: As assessed at the time of diagnosis, surgical treatment of oligometastases should aim at R0 curativity on the entire disease volume, including both the primary tumor and its metastases. Conversion surgery was defined as surgery on the residual volume of disease, which was initially not resectable for technical and/or oncological reasons but nevertheless responded to first-line treatment.

Th17-associated cytokines IL-17 and IL-23 in inflamed skin of Darier disease patients as potential therapeutic targets.

Darier disease (DD) is a rare, inherited multi-organ disorder associated with mutations in the ATP2A2 gene. DD patients often have skin involvement characterized by malodorous, inflamed skin and recurrent, severe infections. Therapeutic options are limited and inadequate for the long-term management of this chronic disease. The aim of this study was to characterize the cutaneous immune infiltrate in DD skin lesions in detail and to identify new therapeutic targets. Using gene and protein expression profiling assays including scRNA sequencing, we demonstrate enhanced expression of Th17-related genes and cytokines and increased numbers of Th17 cells in six DD patients. We provide evidence that targeting the IL-17/IL-23 axis in a case series of three DD patients with monoclonal antibodies is efficacious with significant clinical improvement. As DD is a chronic, relapsing disease, our findings might pave the way toward additional options for the long-term management of skin inflammation in patients with DD.

Transformer-based biomarker prediction from colorectal cancer histology: A large-scale multicentric study.

Deep learning (DL) can accelerate the prediction of prognostic biomarkers from routine pathology slides in colorectal cancer (CRC). However, current approaches rely on convolutional neural networks (CNNs) and have mostly been validated on small patient cohorts. Here, we develop a new transformer-based pipeline for end-to-end biomarker prediction from pathology slides by combining a pre-trained transformer encoder with a transformer network for patch aggregation. Our transformer-based approach substantially improves the performance, generalizability, data efficiency, and interpretability as compared with current state-of-the-art algorithms. After training and evaluating on a large multicenter cohort of over 13,000 patients from 16 colorectal cancer cohorts, we achieve a sensitivity of 0.99 with a negative predictive value of over 0.99 for prediction of microsatellite instability (MSI) on surgical resection specimens. We demonstrate that resection specimen-only training reaches clinical-grade performance on endoscopic biopsy tissue, solving a long-standing diagnostic problem.

Minimally invasive autopsies for the investigation of pulmonary pathology of COVID-19-experiences of a longitudinal series of 92 patients.

Minimally invasive autopsies (MIAs) allow the collection of tissue samples for diagnostic and research purposes in special situations, e.g., when there is a high risk of infection which is the case in the context of COVID-19 or restrictions due to legal or personal reasons. We performed MIA to analyze lung tissue from 92 COVID-19 patients (mean age 78 years; range 48-98; 35 women, 57 men), representing 44% of all patients who died from the disease between October 2020 and April 2021. An intercostal approach was used with removal of a 5-cm rib section followed by manual collection of four lung tissue samples (5-8 cm in size). Diffuse alveolar damage (DAD) was found in 89 (97%) patients at various stages. Exudative DAD (eDAD) predominated in 18 (20%) patients, proliferative DAD (pDAD) in 43 (47%) patients, and mixed DAD (mDAD) in 31 (34%) patients. There were no significant differences in the predominant DAD pattern between tissue samples from the same patient. Additional purulent components were present in 46 (50%) cases. Fungi were detected in 11 (12%) patients. The pDAD pattern was associated with longer hospital stay including intensive care unit (p=0.026 and p<0.001) and younger age (p=0.019). Positive bronchoalveolar lavage and blood cultures were observed more frequently in pDAD patterns (p<0.001; p=0.018). In contrast, there was no significant association between intravital positive microbiological results and superimposed bronchopneumonia or fungal infection at autopsy. Having demonstrated the characteristic lung changes in a large longitudinal autopsy series, we conclude that the presented MIA approach can be considered a reliable and safe method for performing post mortem lung diagnostics in COVID-19 and other high-risk situations. The lack of correlation between histological changes indicative of bacterial or fungal superinfection and microbiology could have clinical implications for disease and treatment surveillance.

Programmed death-ligand 1 (PD-L1) expression in primary gastric adenocarcinoma and matched metastases.

BACKGROUND: Combination of immunotherapy and chemotherapy is recommended for first line treatment of gastric adenocarcinoma (GC) patients with locally advanced unresectable disease or metastatic disease. However, data regarding the concordance rate between PD-L1 combined positive score (CPS) in primary GC and matched regional lymph node metastasis (LNmet) or matched distant metastasis (Dmet) is limited. METHODS: Tissue microarray sections from primary resected GC, LNmet and Dmet were immunohistochemically stained with anti-PD-L1 (clone SP263). PD-L1 expression was scored separately in tumour cells and immune cells and compared between matched primary GC, LNmet and/or Dmet. CPS was calculated and results for CPS cut-offs 1 and 5 were compared between matched samples. RESULTS: 275 PD-L1 stained GC were analysed. 189 primary GC had matched LNmet. CPS cut-off 1 concordance rate between primary GC and LNmet was 77%. 23 primary GC had matched Dmet but no matched LNmet, CPS cut-off 1 concordance rate was 70%. 63 primary GC had both matched LNmet and matched Dmet, CPS cut-off 1 concordance rate of 67%. CPS cut-off 5 results were similar. The proportion of PD-L1 positive tumour cells increased from primary GC (26%) to LNmet (42%) and was highest in Dmet (75%). CONCLUSION: Our study showed up to 33% discordance of PD-L1 CPS between primary GC and LNmet and/or Dmet suggesting that multiple biopsies of primary GC and metastatic sites might need to be tested before considering treatment options. Moreover, this is the first study that seems to suggest that tumour cells acquire PD-L1 expression during disease progression.

Tumour infiltrating lymphocytes and survival after adjuvant chemotherapy in patients with gastric cancer: post-hoc analysis of the CLASSIC trial.

BACKGROUND: Only a subset of gastric cancer (GC) patients with stage II-III benefits from chemotherapy after surgery. Tumour infiltrating lymphocytes per area (TIL density) has been suggested as a potential predictive biomarker of chemotherapy benefit. METHODS: We quantified TIL density in digital images of haematoxylin-eosin (HE) stained tissue using deep learning in 307 GC patients of the Yonsei Cancer Center (YCC) (193 surgery+adjuvant chemotherapy [S + C], 114 surgery alone [S]) and 629 CLASSIC trial GC patients (325 S + C and 304 S). The relationship between TIL density, disease-free survival (DFS) and clinicopathological variables was analysed. RESULTS: YCC S patients and CLASSIC S patients with high TIL density had longer DFS than S patients with low TIL density (P = 0.007 and P = 0.013, respectively). Furthermore, CLASSIC patients with low TIL density had longer DFS if treated with S + C compared to S (P = 0.003). No significant relationship of TIL density with other clinicopathological variables was found. CONCLUSION: This is the first study to suggest TIL density automatically quantified in routine HE stained tissue sections as a novel, clinically useful biomarker to identify stage II-III GC patients deriving benefit from adjuvant chemotherapy. Validation of our results in a prospective study is warranted.

Pathological regression of primary tumour and metastatic lymph nodes following chemotherapy in resectable OG cancer: pooled analysis of two trials.

BACKGROUND: No definitive largescale data exist evaluating the role of pathologically defined regression changes within the primary tumour and lymph nodes (LN) of resected oesophagogastric (OG) adenocarcinoma following neoadjuvant chemotherapy and the impact on survival. METHODS: Data and samples from two large prospective randomised trials (UK MRC OE05 and ST03) were pooled. Stained slides were available for central pathology review from 1619 patients. Mandard tumour regression grade (TRG) and regression of tumour within LNs (LNR: scored as present/absent) were assessed and correlated with overall survival (OS) using a Cox regression model. An exploratory analysis to define subgroups with distinct prognoses was conducted using a classification and regression tree (CART) analysis. RESULTS: Neither trial demonstrated a relationship between TRG score and the presence or absence of LNR. In univariable analysis, lower TRG, lower ypN stage, lower ypT stage, presence of LNR, presence of well/moderate tumour differentiation, and absence of tumour at resection margin were all associated with better OS. However, the multivariable analysis demonstrated that only ypN, ypT, grade of differentiation and resection margin (R0) were independent indicators of prognosis. Exploratory CART analysis identified six subgroups with 3-year OS ranging from 83% to 22%; with ypN stage being the most important single prognostic variable. CONCLUSIONS: Pathological LN stage within the resection specimen was the single most important determiner of survival. Our results suggest that the assessment of regression changes within the primary tumour or LNs may not be necessary to define the prognosis further.

Direct prediction of genetic aberrations from pathology images in gastric cancer with swarm learning.

BACKGROUND: Computational pathology uses deep learning (DL) to extract biomarkers from routine pathology slides. Large multicentric datasets improve performance, but such datasets are scarce for gastric cancer. This limitation could be overcome by Swarm Learning (SL). METHODS: Here, we report the results of a multicentric retrospective study of SL for prediction of molecular biomarkers in gastric cancer. We collected tissue samples with known microsatellite instability (MSI) and Epstein-Barr Virus (EBV) status from four patient cohorts from Switzerland, Germany, the UK and the USA, storing each dataset on a physically separate computer. RESULTS: On an external validation cohort, the SL-based classifier reached an area under the receiver operating curve (AUROC) of 0.8092 (± 0.0132) for MSI prediction and 0.8372 (± 0.0179) for EBV prediction. The centralized model, which was trained on all datasets on a single computer, reached a similar performance. CONCLUSIONS: Our findings demonstrate the feasibility of SL-based molecular biomarkers in gastric cancer. In the future, SL could be used for collaborative training and, thus, improve the performance of these biomarkers. This may ultimately result in clinical-grade performance and generalizability.

Surgical pathology of adenocarcinomas arising around or within the gastroesophageal junction.

Classification of adenocarcinomas (AC) arising around or within the gastroesophageal junction (GEJ) is hampered by major morphologic and phenotypic overlaps. We reviewed the surgical pathology of esophagectomy specimens of 115 primary resected AC of the esophagus as defined by the 5th edition of the WHO classification regarding the anatomical site of the tumor, with corresponding categorization according to the Siewert AEG Classification and the preceding 4th edition of the WHO (discriminating esophageal adenocarcinomas/EAC and adenocarcinomas of the gastroesophageal junction/AdGEJ), and further histology findings. In addition, immunohistochemistry (IHC) for CDX2, CK7, CK20, MUC2, MUC5AC and MUC6 was performed. Sixty-eight cases were Siewert AEG type I and 47 cases Siewert AEG type II. Out of the AEG I tumors, 26 were classified as AdGEJ. Regardless of the classification system, more proximally located tumors showed less aggressive behavior with lower rates of lymph node metastases, lymphatic, venous and perineural invasion, better histological differentiation (p < 0.05 each) and were more frequently associated with pre-neoplastic Barrett's mucosa (p < 0.001). Histologically, the tumors displayed intestinal morphology in the majority of cases. IHC showed non-conclusive patterns with a frequent CK7+/CK20+ immunophenotype in all tumors, but also a gastric MUC5AC+ and MUC6+ phenotype in some proximal tumors. In conclusion, histology of the tumors and IHC failed to distinguish reliably between more proximal and more distal tumors. The presence of Barrett's mucosa rather than location alone, however, may help to further differentiating adenocarcinomas arising in this region and may be indicative for a particular biologic type.

UICC Staging after Neoadjuvant/Perioperative Chemotherapy Reveals No Significant Survival Differences Compared to Primary Surgery for Locally Advanced Gastric Cancer.

Background: The applicability of UICC TNM staging for gastric cancer (GC) patients treated with neoadjuvant chemotherapy (nCTX) and surgery was not yet analyzed in comparison to patients undergoing primary surgery (PS). The purpose of this analysis was to analyze if the prognostic impact of TNM staging after nCTx is comparable with PS. Methods: Data for patients having been treated for GC with or without nCTx between 1990 and 2016 were analyzed. Uni-(URA) and multivariable regression analyses (MRA) were performed to identify predictors. Survival according to the UICC 8th edition stages was analyzed by the Kaplan−Meier method and cox regression analysis. Propensity score matching (PSM) was performed to balance for confounders. Results: 1149 patients with GC were eligible for primary analysis. URA demonstrated age (p < 0.0001), tumor localization (p < 0.0001), clinical UICC-stage, complications, UICC stage 0, IIB-IIIC, Lauren subtype, grading, and R-stage to be significantly associated with OS. MRA revealed that age, distal tumor localization, more than 25 dissected lymph nodes, UICC stage 0, IIB-IIIC, and Lauren subtype were significantly and independently related to OS. After PSM, survival analyses revealed only a significant difference for pN2/ypN2 (p = 0.03), while all other T and N stages were comparable. Conclusion: UICC dependent survival stages do not change significantly after nCTx treatment for GC. Therefore, UICC staging in its present version is applicable to patients undergoing nCTx.

A 25 mm Circular Stapler Anastomosis Is Associated with Higher Anastomotic Leakage Rates Following Minimally Invasive Ivor Lewis Operation.

(1) Background: Minimally invasive oesophagectomy (MIE) with intrathoracic anastomosis is increasingly used in treating patients with oesophageal cancer. Anastomotic leakage (AL) remains a critical perioperative complication, despite recent advances in surgical techniques. It remains unclear to what extent the size of the circular stapler (CS), a 25 mm CS or a bigger CS, may affect the incidence of AL. This study aimed to evaluate whether the CS size in oesophagogastrostomy affects the postoperative AL rates and related morbidity in MIE. (2) Methods: We conducted a retrospective review of consecutive patients who had undergone thoracic MIE between August 2014 and July 2019 using a CS oesophagogastric anastomosis at the level of the Vena azygos. The patients were grouped according to CS size (mm): small-sized (SS25) and large-sized (LS29). The patient demographics, data regarding morbidity, and clinical outcomes were compared. The primary outcome measure was the AL rate related to the stapler size. (3) Results: A total of 119 patients were included (SS25: n = 65; LS29: n = 54). Except for the distribution of squamous cell carcinoma, the demographics were similar in each group. The AL rate was 3.7% in the LS29 group and 18.5% in the SS25 group (p = 0.01). The major morbidity (CD ≥ 3a) was significantly more frequent in the SS25 group compared with the LS29 group (p = 0.02). CS size, pulmonary complications, and cardiovascular disease were independent risk factors for AL in the multivariate analysis. (4) Conclusions: A 29 mm CS is associated with significantly improved surgical outcomes following standard MIE at the level of the azygos vein and should be conducted whenever technically feasible.

Autonomic and circulatory alterations persist despite adequate resuscitation in a 5-day sepsis swine experiment.

Autonomic and vascular failures are common phenotypes of sepsis, typically characterized by tachycardia despite corrected hypotension/hypovolemia, vasopressor resistance, increased arterial stiffness and decreased peripheral vascular resistance. In a 5-day swine experiment of polymicrobial sepsis we aimed at characterizing arterial properties and autonomic mechanisms responsible for cardiovascular homeostasis regulation, with the final goal to verify whether the resuscitation therapy in agreement with standard guidelines was successful in restoring a physiological condition of hemodynamic profile, cardiovascular interactions and autonomic control. Twenty pigs were randomized to polymicrobial sepsis and protocol-based resuscitation or to prolonged mechanical ventilation and sedation without sepsis. The animals were studied at baseline, after sepsis development, and every 24 h during the 3-days resuscitation period. Beat-to-beat carotid blood pressure (BP), carotid blood flow, and central venous pressure were continuously recorded. The two-element Windkessel model was adopted to study carotid arterial compliance, systemic vascular resistance and characteristic time constant τ. Effective arterial elastance was calculated as a simple estimate of total arterial load. Cardiac baroreflex sensitivity (BRS) and low frequency (LF) spectral power of diastolic BP were computed to assess autonomic activity. Sepsis induced significant vascular and autonomic alterations, manifested as increased arterial stiffness, decreased vascular resistance and τ constant, reduced BRS and LF power, higher arterial afterload and elevated heart rate in septic pigs compared to sham animals. This compromised condition was persistent until the end of the experiment, despite achievement of recommended resuscitation goals by administered vasopressors and fluids. Vascular and autonomic alterations persist 3 days after goal-directed resuscitation in a clinically relevant sepsis model. We hypothesize that the addition of these variables to standard clinical markers may better profile patients' response to treatment and this could drive a more tailored therapy which could have a potential impact on long-term outcomes.

Adversarial attacks and adversarial robustness in computational pathology.

Artificial Intelligence (AI) can support diagnostic workflows in oncology by aiding diagnosis and providing biomarkers directly from routine pathology slides. However, AI applications are vulnerable to adversarial attacks. Hence, it is essential to quantify and mitigate this risk before widespread clinical use. Here, we show that convolutional neural networks (CNNs) are highly susceptible to white- and black-box adversarial attacks in clinically relevant weakly-supervised classification tasks. Adversarially robust training and dual batch normalization (DBN) are possible mitigation strategies but require precise knowledge of the type of attack used in the inference. We demonstrate that vision transformers (ViTs) perform equally well compared to CNNs at baseline, but are orders of magnitude more robust to white- and black-box attacks. At a mechanistic level, we show that this is associated with a more robust latent representation of clinically relevant categories in ViTs compared to CNNs. Our results are in line with previous theoretical studies and provide empirical evidence that ViTs are robust learners in computational pathology. This implies that large-scale rollout of AI models in computational pathology should rely on ViTs rather than CNN-based classifiers to provide inherent protection against perturbation of the input data, especially adversarial attacks.

Hepatic blood flow regulation but not oxygen extraction capability is impaired in prolonged experimental abdominal sepsis.

Impaired oxygen utilization has been proposed to play a significant role in sepsis-induced liver dysfunction, but its magnitude and temporal course during prolonged resuscitation is controversial. The aim of this study is to evaluate the capability of the liver to increase oxygen extraction in sepsis during repeated acute portal vein blood flow reduction. Twenty anesthetized and mechanically ventilated pigs with hepatic hemodynamic monitoring were randomized to fecal peritonitis or controls (n = 10, each). After 8-h untreated sepsis, the animals were resuscitated for three days. The ability to increase hepatic O2 extraction was evaluated by repeated, acute decreases in hepatic oxygen delivery (Do2) via reduction of portal flow. Blood samples for liver function and liver biopsies were obtained repeatedly. Although liver function tests, ATP content, and Do2 remained unaltered, there were signs of liver injury in blood samples and overt liver cell necrosis in biopsies. With acute portal vein occlusion, hepatic Do2 decreased more in septic animals compared with controls [max. decrease: 1.66 ± 0.68 mL/min/kg in sepsis vs. 1.19 ± 0.42 mL/min/kg in controls; portal venous flow (Qpv) reduction-sepsis interaction: P = 0.028]. Hepatic arterial buffer response (HABR) was impaired but recovered after 3-day resuscitation, whereas hepatic oxygen extraction increased similarly during the procedures in both groups (max. increase: 0.27 ± 0.13 in sepsis vs. 0.18 ± 0.09 in controls; all P > 0.05). Our data indicate maintained capacity of the liver to acutely increase O2 extraction, whereas blood flow regulation is transiently impaired with the potential to contribute to liver injury in sepsis.NEW & NOTEWORTHY The capacity to acutely increase hepatic O2 extraction with portal flow reduction is maintained in sepsis with accompanying liver injury, but hepatic blood flow regulation is impaired.

Neoadjuvant Radiation in High-Grade Soft-Tissue Sarcomas: Histopathologic Features and Response Evaluation.

In this study, we sought to determine the prognostic value of both the European Organization for Research and Treatment of Cancer-Soft Tissue and Bone Sarcoma Group (EORTC-STBSG) score and the histologic parameters viable tumor, coagulative necrosis, hyalinization/fibrosis, and infarction in patients (n=64) with localized, nonmetastatic high-grade soft-tissue sarcomas after preoperative radiomonotherapy. A standardized macroscopic workup for pretreated surgical specimen including evaluation of a whole section of high-grade soft tissue sarcomas in the largest diameter, was used. Association with overall survival and disease-free survival was assessed. Limb salvage could be accomplished in 98.4% of patients. Overall, 90.6% tumors had negative resection margins. The median postoperative tumor diameter was 9 cm. Undifferentiated pleomorphic sarcoma (42.2%) and myxofibrosarcoma (17.2%) were the most common diagnoses. In all, 9.4% of patients had local recurrence despite clear resection margins, and 50% had distant metastases. Morphologic mapping suggests an overall heterogenous intratumoral response to radiotherapy, with significant differences among histologic subtypes. Complete regression (0% vital tumor cells) was not seen. Categorizing the results according to the proposed EORTC-STBSG 5-tier response score, <1% viable tumor cells were seen in 3.1%, ≥1% to <10% viable tumor cells in 20.4%, ≥10% to <50% viable tumor cells in 35.9% and ≥50% viable tumor cells in 40.6% of cases. Mean values for viable tumor cells were 40% (range: 1% to 100%), coagulative necrosis 5% (0% to 60%), hyalinization/fibrosis 25% (0% to 90%) and infarction 15% (0% to 79%). Hyalinization/fibrosis was a significant independent prognostic factor for overall survival (hazard ratio=4.4; P =0.047), while the other histologic parameters including the EORTC-STBSG score were not prognostic.

Benchmarking weakly-supervised deep learning pipelines for whole slide classification in computational pathology.

Artificial intelligence (AI) can extract visual information from histopathological slides and yield biological insight and clinical biomarkers. Whole slide images are cut into thousands of tiles and classification problems are often weakly-supervised: the ground truth is only known for the slide, not for every single tile. In classical weakly-supervised analysis pipelines, all tiles inherit the slide label while in multiple-instance learning (MIL), only bags of tiles inherit the label. However, it is still unclear how these widely used but markedly different approaches perform relative to each other. We implemented and systematically compared six methods in six clinically relevant end-to-end prediction tasks using data from N=2980 patients for training with rigorous external validation. We tested three classical weakly-supervised approaches with convolutional neural networks and vision transformers (ViT) and three MIL-based approaches with and without an additional attention module. Our results empirically demonstrate that histological tumor subtyping of renal cell carcinoma is an easy task in which all approaches achieve an area under the receiver operating curve (AUROC) of above 0.9. In contrast, we report significant performance differences for clinically relevant tasks of mutation prediction in colorectal, gastric, and bladder cancer. In these mutation prediction tasks, classical weakly-supervised workflows outperformed MIL-based weakly-supervised methods for mutation prediction, which is surprising given their simplicity. This shows that new end-to-end image analysis pipelines in computational pathology should be compared to classical weakly-supervised methods. Also, these findings motivate the development of new methods which combine the elegant assumptions of MIL with the empirically observed higher performance of classical weakly-supervised approaches. We make all source codes publicly available at https://github.com/KatherLab/HIA, allowing easy application of all methods to any similar task.

Neoadjuvant chemotherapy improves survival in patients with oesophageal mucinous adenocarcinoma: Post-hoc analysis of the UK MRC OE02 and OE05 trials.

BACKGROUND: Adenocarcinoma with more than 50% extracellular mucin is a relatively rare histological subtype of gastrointestinal adenocarcinomas. The clinical impact of extracellular mucin in oesophageal adenocarcinoma (OeAC) has not been investigated in detail. We hypothesised that patients with mucinous OeAC (OeACmucin) do not benefit from neoadjuvant chemotherapy. METHODS: OeAC patients either treated by surgery alone in the OE02 trial (S-patients) or by neoadjuvant chemotherapy followed by surgery (CS-patients) in OE02 or OE05 trials were included. Cancers from 1055 resection specimens (OE02 [test cohort]: 187 CS, 185 S; OE05 [validation cohort]: 683 CS) were classified as either mucinous (more than 50% of the tumour area consists of extracellular mucin, OeACmucin) or non-mucinous adenocarcinoma (OeACnon-mucin). The relationship between histological phenotype, clinicopathological characteristics, survival and treatment was analysed. RESULTS: Overall, 7.3% and 9.6% OeAC were classified as OeACmucin in OE02 and OE05, respectively. In OE02, the frequency of OeACmucin was similar in S and CS-patients. Patients with OeACmucin treated with surgery alone had a poorer overall survival compared with OeACnon-mucin patients (hazard ratio: 2.222, 95% confidence interval: 1.08-4.56, P = 0.025). Patients with OeACmucin treated with neoadjuvant chemotherapy and surgery had similar survival as OeACnon-mucin patients in test and validation cohort. CONCLUSIONS: This is the first study to suggest in a post-hoc analysis of material from two independent phase III clinical trials that the poor survival of patients with mucinous OeAC can be improved by neoadjuvant chemotherapy. Future studies are warranted to identify potential underlying biological, biochemical or pharmacokinetic interactions between extracellular mucin and chemotherapy.