Jon Gunnlaugur Jonasson

Endoscopic submucosal dissection (ESD) for pharyngeal squamous cell carcinoma (SCC) has predominantly been performed in Asia but interest in this procedure is growing in the West. This report details the first known Western application of ESD for oropharyngeal SCC. A Caucasian female with superficial oropharyngeal SCC detected during endoscopic surveillance post-ESD for esophageal SCC underwent successful en-bloc resection without complications. Pathology confirmed SCC in situ with tumor-free margins. This case underscores the importance of mindset to include the pharynx in upper gastrointestinal endoscopy during surveillance within the gastroenterologist's domain and the potential for broader application of ESD in the West.

Background: Barrett’s esophagus (BE) is a known precursor to esophageal adenocarcinoma (EAC). However, reports on incidence and progression-to-neoplasm rates have been very variable and conflicting. The aims of the study were to evaluate the characteristics of BE and its progression to neoplasm in a large homogeneous population. Methods: This was a retrospective population-based study with patients identified from 11 institutions through the databases in two centralized pathology laboratories. Demographics and relevant clinicopathological features were obtained from medical records among patients with a pathologically confirmed BE by the presence of intestinal metaplasia between 2003 and 2022. Results: A total of 1388 patients were identified with BE: 948 were men (69%); the median age at diagnosis was 62 years (IQR, 53–72). The ratio of long-segment BE to short-segment BE was significantly higher in patients ≥ 60 years (1.15, 284/248) than those ≤ 60 years (0.77, 205/265) (p = 0.0025). At BE diagnosis, 9.4% had neoplasms: LGD (n = 65), HGD (n = 16), and EAC (n = 49). Among 1258 non-dysplastic BE (NDBE) patients, 4.6% developed a neoplasm—LGD (n = 35), HGD (n = 8), and EAC (n = 15)—with a median observation-period of 5 years (IQR, 3–7). Overall, 160 cases with neoplasms were diagnosed in this BE cohort; 130 (74%) were present at initial BE diagnosis, and 58 (26%) progressed to neoplasms from NDBE. Conclusions: The ratio of long-segment BE was found to be significantly higher in patients ≥ 60 years. Around 9% of the patients were diagnosed as harboring a neoplasm concomitantly with BE, accounting for approximately 74% of all neoplasms. After a median follow-up of 5 years, about 5% of BE showed dysplastic or malignant progression.

Optical microscopy has long been the gold standard to analyse tissue samples for the diagnostics of various diseases, such as cancer. The current diagnostic workflow is time-consuming and labour-intensive, and manual annotation by a qualified pathologist is needed. With the ever-increasing number of tissue blocks and the complexity of molecular diagnostics, new approaches have been developed as complimentary or alternative solutions for the current workflow, such as digital pathology and mass spectrometry imaging (MSI). This study compares the performance of a digital pathology workflow using deep learning for tissue recognition and an MSI approach utilising shallow learning to annotate formalin-fixed and paraffin-embedded (FFPE) breast cancer tissue microarrays (TMAs). Results show that both deep learning algorithms based on conventional optical images and MSI-based shallow learning can provide automated diagnostics with F1-scores higher than 90%, with the latter intrinsically built on biochemical information that can be used for further analysis.

Salivary gland tumours (SGT) are a vast and heterogenous group of neoplasms. There is a relative lack of comprehensive nationwide epidemiological studies on the subject. The aim of this nationwide analysis was to gain insight into epidemiological traits, such as site, incidence and histological subtypes of SGT in general. Patients diagnosed with a primary SGT between 1986 and 2015 were identified from The Icelandic Cancer Registry and registries from all pathology departments in Iceland. Information on age, sex, tumour location and histology was retrieved from pathology reports. A total of 687 patients were diagnosed with a SGT, 609 (89%) were benign and 78 (11%) malignant. 9% of parotid gland tumours, 22% of submandibular gland tumours and 26% of minor SGT were malignant. The most common malignant tumours were mucoepidermoid carcinoma, acinic cell carcinoma and adenoid cystic carcinoma. The incidence of benign SGT was 4.9 per 100 000 among men and 7.0 per 100 000 among women. The incidence of malignant tumours was 0.59 per 100 000 for men and 0.79 per 100 000 for women. The proportion of malignant SGT is lower than most often reported. Only 10% of parotid gland tumours, 20% of submandibular gland tumours and 25% of minor salivary gland tumours are malignant.

DNA repair of alkylation damage is defective in various cancers. This occurs through somatically acquired inactivation of the MGMT gene in various cancer types, including breast cancers. In addition to MGMT, the two E. coli AlkB homologs ALKBH2 and ALKBH3 have also been linked to direct reversal of alkylation damage. However, it is currently unknown whether ALKBH2 or ALKBH3 are found inactivated in cancer. Methylome datasets (GSE52865, GSE20713, GSE69914), available through Omnibus, were used to determine whether ALKBH2 or ALKBH3 are found inactivated by CpG promoter methylation. TCGA dataset enabled us to then assess the impact of CpG promoter methylation on mRNA expression for both ALKBH2 and ALKBH3. DNA methylation analysis for the ALKBH3 promoter region was carried out by pyrosequencing (PyroMark Q24) in 265 primary breast tumours and 30 proximal normal breast tissue samples along with 8 breast-derived cell lines. ALKBH3 mRNA and protein expression were analysed in cell lines using RT-PCR and Western blotting, respectively. DNA alkylation damage assay was carried out in cell lines based on immunofluorescence and confocal imaging. Data on clinical parameters and survival outcomes in patients were obtained and assessed in relation to ALKBH3 promoter methylation. The ALKBH3 gene, but not ALKBH2, undergoes CpG promoter methylation and transcriptional silencing in breast cancer. We developed a quantitative alkylation DNA damage assay based on immunofluorescence and confocal imaging revealing higher levels of alkylation damage in association with epigenetic inactivation of the ALKBH3 gene (P = 0.029). In our cohort of 265 primary breast cancer, we found 72 cases showing aberrantly high CpG promoter methylation over the ALKBH3 promoter (27%; 72 out of 265). We further show that increasingly higher degree of ALKBH3 promoter methylation is associated with reduced breast-cancer specific survival times in patients. In this analysis, ALKBH3 promoter methylation at >20% CpG methylation was found to be statistically significantly associated with reduced survival (HR = 2.3; P = 0.012). By thresholding at the clinically relevant CpG methylation level (>20%), we find the incidence of ALKBH3 promoter methylation to be 5% (13 out of 265). ALKBH3 is a novel addition to the catalogue of DNA repair genes found inactivated in breast cancer. Our results underscore a link between defective alkylation repair and breast cancer which, additionally, is found in association with poor disease outcome.

Few studies have compared the prognosis and liver-related mortality in patients with NAFLD (nonalcoholic fatty liver disease) and AFLD (alcoholic fatty liver disease). We aimed to investigate the etiology and liver-related mortality of patients with liver biopsy verified fatty liver disease in a population based setting. In this retrospective study, all patients who underwent a liver biopsy 1984–2009 at the National University Hospital of Iceland were identified through a computerized pathological database with the code for fatty liver. Only patients with NAFLD and AFLD were included and medical records reviewed. The patients were linked to the Hospital Discharge Register, the Causes of Death Registry and Centre for Addiction Medicine. A total of 151 had NAFLD and 94 AFLD with median survival of 24 years and 20 years, respectively (p = NS). A total of 10/151 (7%) patients developed cirrhosis in the NAFLD group and 19/94 (20%) in AFLD group (p = 0.03). The most common cause of death in the NAFLD group was cardiovascular disease (48%). Liver disease was the most common cause of death in the AFLD group (36%), whereas liver-related death occurred in 7% of the NAFLD group. The mean liver-related death rate among the general population during the study period was 0.1% of all deaths. There was a significantly worse survival for patients in the AFLD group compared to the NAFLD group after adjusting for gender, calendar year of diagnosis and age at diagnosis (HR 2.16, p = 0.009). The survival for patients with moderate to severe fibrosis was significantly worse than for patients with mild fibrosis after adjusting for gender, calendar year of diagnosis and age at diagnosis (HR 2.09, p = 0.01). Patients with fatty liver disease showed a markedly higher risk of developing liver-related death compared to the general population. The AFLD group had higher liver-related mortality and had a worse survival than the NAFLD group. Patients with more severe fibrosis at baseline showed a worse survival than patients with none or mild fibrosis at baseline.

The aim of this study is to clarify the prognostic importance of several well-known but still debated pathological variables related to the survival of colon cancer patients. The study focuses on the definition and survival carried by the pT4 category and stage II where the presence of high-risk variables may determine whether or not adjuvant chemotherapy is administered. A retrospective nationwide study was carried out including all colon cancer patients that underwent resection in Iceland between 1990 and 2004 (n = 889). All histopathology was reassessed. Cancer-specific survival (CSS) and overall survival were analysed using Kaplan-Meier and Cox regression analysis. In stage II, the five-year CSS for pT4 was 50% (95% CI, 32–69%), which was the lowest survival observed in that stage. In stage III the five-year CSS was 30% (95% CI, 18–41%) and 37% (95% CI, 26–48%) for pT4 and pN2 tumors, respectively. Lymphatic invasion and differentiation had no prognostic value in stage II. The survival associated with pT4a versus pT4b depends on how these categories are defined with regard to Shepherd's local peritoneal involvement (LPI). In the present series, pT4 is a major indicator of poor prognosis in patients with stage II and III colon carcinoma. Four-tiered TNM or Dukes staging systems are insufficient by not taking this variable into account. Only Shepherd's LPI4 and a subgroup of LPI3 (i.e., borderline LPI3/LPI4) should qualify for the pT4a subcategory. The results do not support lymphatic invasion or poor differentiation as high-risk stage II variables.

Available data correlating symptoms of colon cancer patients with the severity of the disease are very limited. In a population-based setting, we correlated information on symptoms of colon cancer patients with several pathological tumor parameters and survival. Information on all patients diagnosed with colon cancer in Iceland in 1995–2004 for this retrospective, population-based study was obtained from the Icelandic Cancer Registry. Information on symptoms of patients and blood hemoglobin was collected from patients' files. Pathological parameters were obtained from a previously performed standardized tumor review. A total of 768 patients entered this study; the median age was 73 years. Tumors in patients presenting at diagnosis with visible blood in stools were significantly more likely to be of lower grade, having pushing border, conspicuous peritumoral lymphocytic infiltration, and lower frequency of vessel invasion. Patients with abdominal pain and anemia were significantly more likely to have vessel invasion. Logistic regression showed that visible blood in stools was significantly associated with protecting pathological factors (OR range 0.38–0.83, p < 0.05). Tumors in patients presenting with abdominal pain were strongly associated with infiltrative margin and scarce peritumoral lymphocytic infiltration (OR = 1.95; 2.18 respectively, p < 0.05). Changes in bowel habits were strongly associated with vessel invasion (OR = 2.03, p < 0.05). Cox regression showed that blood in stools predicted survival (HR = 0.54). In conclusion, visible blood in stools correlates significantly with all the beneficial pathological parameters analyzed and with better survival of patients. Anemia, general symptoms, changes in bowel habits, acute symptoms, and abdominal pain correlate with more aggressive tumor characteristics and adverse outcome for patients.