Stage IV Colorectal Cancer Patients with High Risk Mutation Profiles Survived 16 Months Longer with Individualized Therapies

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.

Detection of Cancer Mutations by Urine Liquid Biopsy as a Potential Tool in the Clinical Management of Bladder Cancer Patients

Simple Summary

The management of bladder cancer faces multiple challenges concerning the diagnostic and follow-up approaches. The standard diagnostic examination comprises invasive cystoscopy. Urine cytology and recently proposed urine-based biomarkers have been unable to replace cystoscopy, thus prompting calls for improvements. Here, we explore urine liquid biopsy to detect cancer mutations and subsequently evaluate the utility of urine as a suitable specimen for diagnosing bladder cancer. Our results show that the analysis of pre- and postoperative urine with a cost-effective 127-gene panel enables the characterization of tumor mutations. These findings provide cumulative evidence in support of the results of previous studies that testing urine for mutations is a useful strategy to complement the clinical management of bladder cancer patients.

Abstract

The standard diagnostic and follow-up examination for bladder cancer is diagnostic cystoscopy, an invasive test that requires compliance for a long period. Urine cytology and recent biomarkers come short of replacing cystoscopy. Urine liquid biopsy promises to solve this problem and potentially allows early detection, evaluation of treatment efficacy, and surveillance. A previous study reached 52–68% sensitivity using small-panel sequencing but could increase sensitivity to 68–83% by adding aneuploidy and promoter mutation detection. Here, we explore whether a large 127-gene panel alone is sufficient to detect tumor mutations in urine from bladder cancer patients. We recruited twelve bladder cancer patients, obtained preoperative and postoperative urine samples, and successfully analyzed samples from eleven patients. In ten patients, we found at least one mutation in bladder-cancer-associated genes, i.e., a promising sensitivity of 91%. In total, we identified 114 variants, of which 90 were predicted as nonbenign, 30% were associated with cancer, and 13% were actionable according to the CIViC database. Sanger sequencing of the patients’ formalin-fixed, paraffin-embedded (FFPE) tumor tissues confirmed the findings. We concluded that incorporating urine liquid biopsy is a promising strategy in the management of bladder cancer patients.

Actuarial Analysis of Survival among Breast Cancer Patients in Lithuania

Breast cancer is the most common cause of mortality due to cancer for women both in Lithuania and worldwide. Chances of survival after diagnosis differ significantly depending on the stage of disease at the time of diagnosis. Extended term periods are required to estimate survival of, e.g., 15–20 years. Moreover, since mortality of the average population changes with time, estimates of survival of cancer patients derived after a long period of observation can become outdated and can be no longer used to estimate survival of patients who were diagnosed later. Therefore, it can be useful to construct analytic functions that describe survival probabilities. Shorter periods of observation can be enough for such construction. We used the data collected by the Lithuanian Cancer Registry for our analysis. We estimated the chances of survival for up to 5 years after patients were diagnosed with breast cancer in Lithuania. Then we found analytic survival functions which best fit the observed data. At the end of this paper, we provided some examples for applications and directions for further research. We used mainly the Kaplan–Meier method for our study.