Understanding the Landmark Lancet Study on Cancer Across Europe
The recent publication in The Lancet Regional Health – Europe has captured significant attention in the academic and medical communities. Titled 'Comorbidities, medication use, and overall survival in eight cancers: a multinational cohort study of 1.7 million patients across Europe,' this research represents a monumental effort in oncology epidemiology. Conducted through a federated analysis of real-world data from 11 electronic health records (EHRs) and cancer registries spanning eight European countries, it provides unprecedented insights into how common cancers are evolving in terms of patient profiles, treatments, and outcomes.
Researchers from prestigious institutions, including the Vall d’Hebron Institute of Oncology (VHIO) in Spain, the University of Oxford in the UK, the University of Edinburgh, Erasmus MC in the Netherlands, and the University of Helsinki in Finland, collaborated under the European Health Data and Evidence Network (EHDEN). This pan-European partnership exemplifies how universities are driving big data initiatives in higher education and research to tackle pressing health challenges. By standardizing data to the Observational Medical Outcomes Partnership Common Data Model (OMOP-CDM)—a globally recognized framework for observational health data—the study ensured privacy-preserving analysis without centralizing sensitive patient information.
The cohort includes 1,796,278 patients aged 18 and older diagnosed with primary malignant cancers between 2000 and 2019. The eight cancers selected—breast, colorectal, head and neck, liver, lung, pancreatic, prostate, and stomach—account for a substantial portion of Europe's cancer burden. Diagnoses peaked in the 60-79 age group, reflecting Europe's aging population, a trend that universities across the continent are studying in relation to healthcare systems.
Methodology: Harnessing Federated Big Data from European Universities
The study's strength lies in its innovative methodology, which leveraged federated learning to aggregate insights from diverse sources without sharing raw data. Databases included primary care EHRs like CPRD GOLD (UK) and SIDIAP (Spain), hospital records from HUS (Finland) and HUVM (Spain), and population-based cancer registries such as the Norwegian Cancer Registry (CRN) and Netherlands Cancer Registry (NCR). Countries represented are Estonia, Finland, Netherlands, Norway, Portugal, Spain, Switzerland, and the UK.
Patients were followed from diagnosis until death, database exit, or study end, with mortality linked to national registries. Comorbidities were assessed anytime prior to diagnosis, while medications focused on the year before. Overall survival (OS) was calculated using age-standardized Kaplan-Meier estimates at 1, 5, and 10 years, stratified by cancer type, sex, age, database, and country. This rigorous approach, developed by interdisciplinary teams at universities like the University of Tartu (Estonia) and University of Porto (Portugal), sets a benchmark for future multinational studies.
Exclusions ensured data quality: prior malignancies (except non-melanoma skin cancer), same-day diagnosis/death, and sparse cohorts. Median follow-up varied from 111 to 3,082 days, providing a robust timeline to observe trends from 2010-2018 within the broader period.
🔬 Comorbidities: Consistent Patterns with Cancer-Specific Signals
One of the study's standout revelations is the consistency of comorbidity profiles across databases, underscoring shared risk factors in Europe's diverse healthcare landscapes. Hypertension emerged as the most prevalent, ranging from 11% in Finland's HUS to 63% in Estonia's UTARTU. Osteoarthritis followed closely at 8-28%.
- Anaemia was notably high in colorectal (9-23%) and stomach cancers (10-34%), potentially signaling early gastrointestinal bleeding or nutritional deficits.
- Chronic obstructive pulmonary disease (COPD, 18-34%) and pneumonia (5-33%) dominated lung cancer cases, linking smoking history and respiratory vulnerability.
- Type 2 diabetes and chronic liver disease spiked in pancreatic (9-35%) and liver cancers (24-77%), highlighting metabolic and hepatic risks.
Primary care databases captured more age-related conditions like hyperlipidaemia, while registries focused on cancer-linked ones. These patterns, analyzed by teams from the University of Girona and University of Seville, suggest prodromal signs that could inform earlier interventions—a key area for academic researchers in preventive oncology.
Medication Use: Insights into Pre-Diagnosis Treatment Landscapes
Pre-diagnosis medication patterns mirrored comorbidities, with higher usage in primary care settings. Acid-related disorder drugs (15-53%), antibacterials (12-55%), and anti-inflammatories (10-51%) topped lists continent-wide.
- Diuretics and obstructive airway drugs were elevated in liver and lung cancers, respectively.
- Renin-angiotensin agents and lipid-modifying therapies prevailed in prostate cancer (6-47% and 5-38%).
- Antithrombotics featured prominently in colorectal and head/neck cancers.
Psycholeptics (2-37%) indicated mental health burdens. This data, harmonized via OMOP-CDM at institutions like Erasmus MC, reveals how routine prescribing could flag risks, aiding pharmacoepidemiology programs at European universities.
Survival Outcomes: From Optimism in Breast Cancer to Challenges in Pancreatic
Five-year OS varied markedly: breast (76-85%) and prostate (75-83%) excelled, colorectal (56-68%) and head/neck (51-62%) middled, while liver (16-28%), lung (16-25%), pancreatic (3-25%), and stomach (17-39%) lagged. Females fared better in most, except liver/pancreatic/stomach.
| Cancer Type | 5-Year OS Range |
|---|---|
| Breast | 76% (ECi) - 85% (IMASIS) |
| Prostate | 75% (HUVM) - 83% (SIDIAP) |
| Pancreatic | 3% (NCR) - 25% (IMASIS) |
Spanish sources often showed superior rates, possibly due to screening efficacy. University of Oxford statisticians' Kaplan-Meier curves highlight these disparities, urging targeted improvements.
Country and Database Variations: A Mirror to Healthcare Systems
Survival differences underscore systemic variances: UK/Netherlands/Norway lower, Spain higher. Primary care like SIDIAP outperformed registries. For pancreatic cancer, gaps were starkest (3-25%). Federated analysis by University of Helsinki and others proves invaluable for cross-border benchmarking.Read the full Lancet paper
Academic Collaboration: Universities Powering European Cancer Research
This study showcases higher education's pivotal role. Teams from Universitat Autònoma de Barcelona, University of Geneva, and more pooled expertise in data science, oncology, and epidemiology. Such collaborations, supported by EHDEN, train next-gen researchers—opportunities abound in higher ed research jobs.
Implications for Clinical Practice and Policy
Prodromal comorbidities/meds offer early detection cues. Policymakers can address survival gaps via equitable access. Universities advocate for integrated care, aligning with EU's Beating Cancer Plan.
Future Directions: Expanding Real-World Evidence
Limitations like missing staging/smoking data noted; future work could incorporate genomics. Ongoing university-led extensions promise deeper insights into cancer evolution amid aging Europe.
- Incorporate AI for prediction models.
- Expand to more cancers/countries.
- Link to lifestyle data for prevention.
Explore postdoc opportunities in oncology data science.
Patient and Societal Impact
With Europe's cancer cases projected to rise, this study informs personalized medicine. Patients benefit from highlighted risks; academics drive solutions. Check Rate My Professor for top oncology faculty.PubMed abstract
Photo by Waldo Malan on Unsplash
Conclusion: A Call to Action for European Academia
This Lancet study illuminates cancer's multifaceted evolution, urging sustained university investment. For career growth, visit higher-ed-jobs, university-jobs, higher-ed-career-advice, and rate-my-professor.
