Epidemiological Trends of Haematological Malignancies in Belgium 2004-2018: Older Patients Show the Greatest Improvement in Survival.

(1) Background: Haematological malignancies (HMs) represent a heterogeneous group of mostly rare cancers that differ in pathophysiology, incidence, and outcome. (2) Methods: Our study aims to understand the epidemiological situation and trends of 24 main types of HMs in Belgium over a 15-year period, with a focus on the impact of age. Age-standardised incidence, average annual percentage change (AAPC), 5- and 10-year relative survival (RS) and RS trends were estimated for all HMs (N = 94,415) diagnosed between 2004 and 2018. (3) Results: Incidence rates of HM increased, mainly in the 70+ age group (AAPC: 3%). RS varied by age and HM type. For each HM type, outcome decreased with age. The greatest decrease with age in 5-year RS is observed for aggressive HM, acute myeloid leukaemia (AML), acute lymphoblastic leukaemia, and Burkitt lymphoma, from 67%, 90%, and 97% below 20 years, to 2%, 12%, and 16% above 80 years of age, respectively. The moderate improvement in 5-year RS over the 2004-2018 period for all HMs, of +5 percentage point (pp), masks highly heterogenous outcomes by HM type and age group. The most impressive improvements are observed in the 80+ group: +45, +33, +28, and +16 pp for Hodgkin lymphoma, immunoproliferative disorders, follicular lymphoma, and chronic myeloid leukaemia, respectively. (4) Conclusions: The increasing incidence and survival over the 2004-2018 period are likely explained by diagnostic and therapeutic innovations, which have spread to populations not targeted by clinical trials, especially older adults. This real-world population-based study highlights entities that need significant improvement, such as AML.

Machine Learning Algorithm to Estimate Distant Breast Cancer Recurrence at the Population Level with Administrative Data.

Purpose: High-quality population-based cancer recurrence data are scarcely available, mainly due to complexity and cost of registration. For the first time in Belgium, we developed a tool to estimate distant recurrence after a breast cancer diagnosis at the population level, based on real-world cancer registration and administrative data. Methods: Data on distant cancer recurrence (including progression) from patients diagnosed with breast cancer between 2009-2014 were collected from medical files at 9 Belgian centers to train, test and externally validate an algorithm (i.e., gold standard). Distant recurrence was defined as the occurrence of distant metastases between 120 days and within 10 years after the primary diagnosis, with follow-up until December 31, 2018. Data from the gold standard were linked to population-based data from the Belgian Cancer Registry (BCR) and administrative data sources. Potential features to detect recurrences in administrative data were defined based on expert opinion from breast oncologists, and subsequently selected using bootstrap aggregation. Based on the selected features, classification and regression tree (CART) analysis was performed to construct an algorithm for classifying patients as having a distant recurrence or not. Results: A total of 2507 patients were included of whom 216 had a distant recurrence in the clinical data set. The performance of the algorithm showed sensitivity of 79.5% (95% CI 68.8-87.8%), positive predictive value (PPV) of 79.5% (95% CI 68.8-87.8%), and accuracy of 96.7% (95% CI 95.4-97.7%). The external validation resulted in a sensitivity of 84.1% (95% CI 74.4-91.3%), PPV of 84.1% (95% CI 74.4-91.3%), and an accuracy of 96.8% (95% CI 95.4-97.9%). Conclusion: Our algorithm detected distant breast cancer recurrences with an overall good accuracy of 96.8% for patients with breast cancer, as observed in the first multi-centric external validation exercise.

Impact of the COVID-19 Pandemic on Incidence and Observed Survival of Malignant Brain Tumors in Belgium.

(1) Background: This study evaluates the impact of the COVID-19 pandemic on the incidence, treatment, and survival of adults diagnosed with malignant brain tumors in Belgium in 2020. (2) Methods: We examined patients aged 20 and older with malignant brain tumors (2004-2020) from the Belgian Cancer Registry database, assessing incidence, WHO performance status, vital status, and treatment data. We compared 2020 incidence rates with projected rates and age-standardized rates to 2015-2019. The Kaplan-Meier method was used to assess observed survival (OS). (3) Results: In 2020, there was an 8% drop in age-specific incidence rates, particularly for those over 50. Incidence rates plunged by 37% in April 2020 during the first COVID-19 peak but partially recovered by July. For all malignant brain tumors together, the two-year OS decreased by four percentage points (p.p.) in 2020 and three p.p. in 2019, compared to that in 2015-2018. Fewer patients (-9 p.p.) with glioblastoma underwent surgery, and the proportion of patients not receiving surgery, radiotherapy, or systemic therapy increased by six percentage points in 2020. (4) Conclusions: The COVID-19 pandemic profoundly impacted the diagnosis, treatment strategies, and survival of brain tumor patients in Belgium during 2020. These findings should guide policymakers in future outbreak responses, emphasizing the need to maintain or adapt (neuro)-oncological care pathways and promote informed decision making when care capacity is limited.

Machine Learning-Based Extraction of Breast Cancer Receptor Status From Bilingual Free-Text Pathology Reports.

As part of its core business of gathering population-based information on new cancer diagnoses, the Belgian Cancer Registry receives free-text pathology reports, describing results of (pre-)malignant specimens. These reports are provided by 82 laboratories and written in 2 national languages, Dutch or French. For breast cancer, the reports characterize the status of estrogen receptor, progesterone receptor, and Erb-b2 receptor tyrosine kinase 2. These biomarkers are related with tumor growth and prognosis and are essential to define therapeutic management. The availability of population-scale information about their status in breast cancer patients can therefore be considered crucial to enrich real-world scientific studies and to guide public health policies regarding personalized medicine. The main objective of this study is to expand the data available at the Belgian Cancer Registry by automatically extracting the status of these biomarkers from the pathology reports. Various types of numeric features are computed from over 1,300 manually annotated reports linked to breast tumors diagnosed in 2014. A range of popular machine learning classifiers, such as support vector machines, random forests and logistic regressions, are trained on this data and compared using their F 1 scores on a separate validation set. On a held-out test set, the best performing classifiers achieve F 1 scores ranging from 0.89 to 0.92 for the four classification tasks. The extraction is thus reliable and allows to significantly increase the availability of this valuable information on breast cancer receptor status at a population level.

Cause of death for patients with breast cancer: discordance between death certificates and medical files, and impact on survival estimates.

BACKGROUND: Registration and coding of cause of death is prone to error since determining the exact underlying condition leading directly to death is challenging. In this study, causes of death from the death certificates were compared to patients' medical files interpreted by experts at University Hospitals Leuven (UHL), to assess concordance between sources and its impact on cancer survival assessment. METHODS: Breast cancer patients treated at UHL (2009-2014) (follow-up until December 31st 2016) were included in this study. Cause of death was obtained from death certificates and expert-reviewed medical files at UHL. Agreement was calculated using Cohen's kappa coefficient. Cause-specific survival (CSS) was calculated using the Kaplan-Meier method and the relative survival probability (RS) using the Ederer II and Pohar Perme method. RESULTS: A total of 2862 patients, of whom 354 died, were included. We found an agreement of 84.7% (kappa-value of 0.69 (95% C.I.: 0.62-0.77)) between death certificates and medical files. Death certificates had 10.7% false positive and 4.5% false negative rates. However, five-year CSS and RS measures were comparable for both sources. CONCLUSION: For breast cancer patients included in our study, fair agreement of cause of death was seen between death certificates and medical files with similar CSS and RS estimations.

A Systematic Review of Estimating Breast Cancer Recurrence at the Population Level With Administrative Data.

BACKGROUND: Exact numbers of breast cancer recurrences are currently unknown at the population level, because they are challenging to actively collect. Previously, real-world data such as administrative claims have been used within expert- or data-driven (machine learning) algorithms for estimating cancer recurrence. We present the first systematic review and meta-analysis, to our knowledge, of publications estimating breast cancer recurrence at the population level using algorithms based on administrative data. METHODS: The systematic literature search followed Preferred Reporting Items for Systematic Reviews and Meta-Analysis guidelines. We evaluated and compared sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy of algorithms. A random-effects meta-analysis was performed using a generalized linear mixed model to obtain a pooled estimate of accuracy. RESULTS: Seventeen articles met the inclusion criteria. Most articles used information from medical files as the gold standard, defined as any recurrence. Two studies included bone metastases only in the definition of recurrence. Fewer studies used a model-based approach (decision trees or logistic regression) (41.2%) compared with studies using detection rules without specified model (58.8%). The generalized linear mixed model for all recurrence types reported an accuracy of 92.2% (95% confidence interval = 88.4% to 94.8%). CONCLUSIONS: Publications reporting algorithms for detecting breast cancer recurrence are limited in number and heterogeneous. A thorough analysis of the existing algorithms demonstrated the need for more standardization and validation. The meta-analysis reported a high accuracy overall, which indicates algorithms as promising tools to identify breast cancer recurrence at the population level. The rule-based approach combined with emerging machine learning algorithms could be interesting to explore in the future.

Interleukin-6 dynamics as a basis for an early-warning monitor for sepsis and inflammation in individual pigs.

Static interleukin-6 (IL-6) levels of pigs contain considerable individual differences, which obstruct the practical use of IL-6 for disease monitoring purposes. It was hypothesised that interleukin-6 (IL-6) dynamics could be used to quantify these individual differences and carries critical information of the individual pig infection status. Time series of IL-6 responses in 25 pigs were analysed before and after infection by Actinobacillus pleuropneumoniae. The results indicated that amplitude increases of IL-6 fluctuations of individual pigs rather than static IL-6 values should be used as indicator of the infection state. This study shows the added value for IL-6 time series analyses of individual pigs. These results are a first step towards the development of objective individualised methods for monitoring and early detection of sepsis and inflammation processes in pigs by integrating animal response dynamics.

Rethinking food anticipatory activity in the activity-based anorexia rat model.

When a rat is on a limited fixed-time food schedule with full access to a running wheel (activity-based anorexia model, ABA), its activity level will increase hours prior to the feeding period. This activity, called food-anticipatory activity (FAA), is a hypothesized parallel to the hyperactivity symptom in human anorexia nervosa. To investigate in depth the characteristics of FAA, we retrospectively analyzed the level of FAA and activities during other periods in ABA rats. To our surprise, rats with the most body weight loss have the lowest level of FAA, which contradicts the previously established link between FAA and the severity of ABA symptoms. On the contrary, our study shows that postprandial activities are more directly related to weight loss. We conclude that FAA alone may not be sufficient to reflect model severity, and activities during other periods may be of potential value in studies using ABA model.