The last five years have seen an explosion of interest in real-world data (RWD) for oncology, and nowhere is this more evident than in the scientific literature. Searching PubMed for “real world data oncology” yields 12,000+ publications dated 2020–2025, accounting for well over 90% of all such records since the 1980s [1]. Review articles alone contribute about 2,168 papers in this period [1,2]. A review article summarises, synthesises and critically analyses existing research on a specific topic, providing a comprehensive overview of the current state of knowledge rather than presenting new experimental data; common subtypes include narrative literature reviews, systematic reviews and meta-analyses. This surge signals a paradigm shift: clinicians, researchers and regulators are embracing data from everyday practice to complement traditional clinical trials.

What are researchers studying?

A closer look at highly cited papers shows that real-world data (RWD) oncology research clusters into three broad themes, each with distinctive implications:

1) Strategic and policy-driven analyses

A substantial body of reviews, guidance, and position pieces focuses on standards for RWD/RWE quality, reporting, and regulatory use, as well as system-level barriers (reimbursement, access, harmonisation). Representative examples include an ESMO-affiliated review on reporting and use of RWE in colorectal cancer and the role of pragmatic/registry-based RCTs; a European “call to action” on embedding biomarker-informed precision oncology into practice and policy; and data-quality frameworks for EHR-derived oncology RWD. Taken together, these pieces argue for robust curation, transparent methods and standardised reporting to support decision-making [3-5].

2) Disease-specific and regional real-world analyses

A large share of publications are observational studies using registries, claims or EHR cohorts to map utilisation, outcomes and safety across tumour types and settings. Authoritative reviews emphasise that while such evidence can inform care rapidly, pragmatic/registry-based RCTs remain essential for causal inference and generalisability. Illustrative RWD studies span lung-cancer immunotherapy in US community practices, endometrial-cancer outcomes, and broader method/quality appraisals showing variable rigour and the need for stronger designs [6-9].

3) Methodology, AI and evidence generation (fast-growing segment)

Hundreds of papers advance methods for fit-for-purpose RWE – external controls, bias mitigation, and AI/NLP to unlock unstructured clinical text. Recent reviews and studies describe AI for RWD collection/analysis in oncology, clinical NLP to structure free-text notes for care pathways, and ML/NLP pipelines that replicate expert-abstracted EHR analyses at scale [10-13].

Why Sqilline Health leads in RWD structuring and RWE generation

The absolute volume and diversity of RWD oncology publications illustrate both the promise and the complexity of working with real-world sources. Sqilline Health has built its reputation on mastering this complexity. By normalising different data formats, extracting information from unstructured clinical notes via proprietary AI, and applying robust statistical techniques, Sqilline generates real-world evidence (RWE) that meets the evidentiary standards of regulators and payers. The evidence:

  • Built for unstructured text. Most catalogue entries are structured registries or EMR warehouses. Sqilline’s proprietary AI/NLP specialises in extracting variables from free-text (Cyrillic and multilingual) clinical notes, turning narrative care into analysable data [14].
  • Regulatory-grade RWE. The ribociclib study was regulator-required and finalised within the EMA framework [15]. Other Sqilline studies clearly state comparative designs that align RWD with trial evidence (e.g., osimertinib; dabrafenib/trametinib) [16,17].
  • Regional depth, global relevance. Danny operates across multiple countries (BG, RO, HR, RS, SI), adding perspectives from populations under-represented in large trials, key for external validity and equitable evidence [18].

From publications to practice: Sqilline’s EMA RWD Catalogue footprint

  • Ribociclib + letrozole/fulvestrant in advanced/metastatic HR+/HER2- breast cancer – RWE compared with RCT outcomes using Danny [15].
  • Dabrafenib + trametinib in BRAF-positive melanoma – RWE compared with clinical trials using Danny [16].
  • First-line osimertinib in EGFR-mutated NSCLC – RWE benchmarked to clinical trials using Danny [17].
  • Nationwide adherence to breast-cancer guidelines in Bulgaria (Danny listed as the source) [18].
  • Proton therapy for head & neck cancer in routine practice in Bulgaria (Danny listed as the source) [18].

In total, the EMA page for Danny Platform lists seven studies conducted using the data source (five oncology-focused and two in cardiovascular/pulmonary domains) [18]. Against the backdrop of dozens of oncology-related entries added to the catalogue in 2024–2025 (data sources, institutions, networks and studies), Sqilline’s five oncology studies and one registered data source represent a substantial presence in the Oncology “Studies” subset during this period, particularly among Eastern-European real-world analyses [1, 15-18].

The Road Ahead

The rapid growth of RWD oncology research in PubMed underscores an industry-wide pivot toward evidence from routine care. Even if PubMed’s new beta interface reports slightly fewer results – around 10,708 for the same period – the trend is obvious: real-world data is reshaping oncology research. Publications increased from 1,177 (2020) to 2,811 (2024), and strategic discussions, localised analyses and methodological innovations are proliferating [1]. In parallel, the EMA RWD Catalogue shows this evidence becoming operational and regulatory. Sitting at that intersection, Sqilline Health converts complex, unstructured text into regulatory-ready RWE, delivering oncology insights that complement trials and inform decisions.

For Sqilline Health, this represents an enormous opportunity and responsibility: to stay ahead of methodological advances, contribute to best practices, and, above all, transform unstructured clinical text into actionable, real-world evidence that improves cancer care.

References
  1. Search results for “real world data oncology” (results-by-year histogram, 2020–2025). Available from: https://pubmed.ncbi.nlm.nih.gov/?term=real%20world%20data%20oncology (accessed 24 Aug 2025).
  2. “real world data oncology” with filter Publication type: Review (2020–2025). Available from: https://pubmed.ncbi.nlm.nih.gov/?term=real%20world%20data%20oncology&filter=pubt.review&filter=dates.2020-2025 (accessed 24 Aug 2025).
  3. van Nassau SCMW, Bol GM, van der Baan FH, et al. Harnessing the potential of real-world evidence in the treatment of colorectal cancer: where do we stand? Curr Treat Options Oncol. 2024;25(4):405-426. doi:10.1007/s11864-024-01186-4. PMCID: PMC10997699.
  4. Lawler M, Keeling P, Kholmanskikh O, et al. Empowering effective biomarker-driven precision oncology: a call to action. Eur J Cancer. 2024;209:114225. doi:10.1016/j.ejca.2024.114225. PMID: 39053288.
  5. Castellanos EH, Lerro CC, Singal G, et al. Raising the bar for real-world data in oncology. JCO Clin Cancer Inform. 2024;8:e2300231. PMID: 38241599.
  6. Tang M, Cu A, Blinman P, et al. Harnessing real-world evidence to advance cancer research. Curr Oncol. 2023;30(2):143. PMCID: PMC9955401.
  7. Liu SV, Flătaru MC, Stjepanovic N, et al. Real-world outcomes at US oncology practices for first-line pembrolizumab-based therapy in metastatic NSCLC. Cancer Med. 2022;11(24):4690-4702. PMID: 36324586.
  8. Zhang J, Zhang W, Zhang X, et al. Real-world treatment patterns and outcomes in recurrent/advanced endometrial cancer. Eur J Obstet Gynecol Reprod Biol. 2024;292:60-67. PMID: 38569701.
  9. Boyle JM, Kury FSP, Ward RL, et al. Real-world outcomes associated with new cancer medicines approved by FDA/EMA: a retrospective cohort study. Cancer Treat Rev. 2021;99:102257. PMCID: PMC8442759.
  10. Bryant AK, Jin MC, Chen ML, et al. Artificial intelligence to unlock real-world evidence in oncology. ESMO Real-World Data & Digital Oncology. 2024;3(2):100085. PMID: 38899720; PMCID: PMC11187737.
  11. Cerami E, Mahmood U, Nohadani O, et al. Artificial intelligence in oncology: current landscape, challenges, and future directions. Cancer Discov. 2024;14(6):1260-1278. PMCID: PMC11131133.
  12. Lin H, Ni L, Phuong C, Hong JC. Natural language processing for radiation oncology: personalising treatment pathways. Pharmacogenomics Pers Med. 2024;17:65-76. PMCID: PMC10874185.
  13. Benedum CM, Li Y, Zhou K, et al. Replication of real-world evidence in oncology using machine-learning-extracted EHR data. Cancers (Basel). 2023;15(6):1853. doi:10.3390/cancers15061853.
  14. Zhao B. Clinical data extraction and normalization of Cyrillic electronic health records via deep-learning natural language processing. JCO Clin Cancer Inform. 2019;3:1-9. doi:10.1200/CCI.19.00057. PMID: 31577448.
  15. Ribociclib + letrozole/fulvestrant (HR+/HER2- mBC) – HMA-EMA Catalogue. EUPAS1000000083. First published 25 Mar 2024; updated 01 Aug 2025. Available from: https://catalogues.ema.europa.eu/node/3984/administrative-details (accessed 24 Aug 2025).
  16. Dabrafenib + trametinib (BRAF-positive melanoma) – HMA-EMA Catalogue. EUPAS1000000567. First published 06 May 2025; updated 24 Jul 2025. Available from: https://catalogues.ema.europa.eu/node/4468/administrative-details (accessed 24 Aug 2025).
  17. First-line osimertinib (EGFR-mutated NSCLC) – HMA-EMA Catalogue. EUPAS1000000686. First published 29 Jul 2025. Available from: https://catalogues.ema.europa.eu/node/4589/administrative-details (accessed 24 Aug 2025).
  18. Danny Platform – HMA-EMA Catalogues of real-world data sources and studies. First published 01 Feb 2024; updated 17 Oct 2024. Available from: https://catalogues.ema.europa.eu/node/1143/administrative-details (accessed 24 Aug 2025).