Latest Research
All publications from the Cancer3.AI database, newest first.
CA 15.3-driven F-18-FDG PET-CT for early detection of breast cancer recurrence: Diagnostic accuracy and clinical impact.
Neeraj S, et al
A new study published in Bioinformation evaluated whether whole-body F-18-FDG PET-CT scanning could accurately detect hidden breast cancer recurrences in patients whose only warning sign was a rising blood level of the tumor marker CA 15.3. Researchers at Guru Gobind Singh Medical College and Hospital in Faridkot, India, enrolled 50 breast cancer patients under post-treatment surveillance who showed elevated CA 15.3 levels but no clear findings on conventional imaging. The study found a significant correlation between higher CA 15.3 levels and recurrences detected by PET-CT, with bone metastases associated with the highest CA 15.3 values. Critically, PET-CT results led to a change in clinical management in 80% of cases, with most patients receiving targeted radiotherapy, some starting new systemic therapy, and others beginning endocrine therapy. These findings underscore the clinical value of PET-CT as an essential next-step tool when standard imaging fails to explain a rising tumor marker, potentially allowing earlier and more precise treatment of recurrent breast cancer.
Bioinformation
Source →Multimodal AI for pneumonia and lung cancer classification using x-ray and HRCT.
Pradip C, et al
Researchers developed DeepScan, a multimodal artificial intelligence system designed to improve the diagnosis of pneumonia and lung cancer using two types of medical imaging: chest X-rays and high-resolution computed tomography (HRCT) scans. The system combines two deep learning neural networks — ResNet-50 for analyzing X-rays and DenseNet-121 for HRCT — and merges their outputs through a late-fusion architecture to produce a unified diagnostic decision. When tested on 2,000 patients, DeepScan achieved 94.6% accuracy, 95.2% sensitivity, 93.9% specificity, and an area under the curve (AUC) of 0.97, outperforming models that relied on only one imaging type. Critically, the multimodal approach reduced false negatives for early-stage lung cancer and better distinguished cancer from pneumonia, two conditions that can appear similar on imaging alone. These results suggest that combining imaging modalities with AI could support clinicians in making earlier and more accurate diagnoses, potentially improving patient outcomes through timely intervention.
Bioinformation
Source →Circulating extracellular vesicle DNA (EV-DNA) and cell-free DNA in ovarian cancer: pathological relevance and diagnostic applications.
Guanzon D, et al
This review examines the diagnostic potential of liquid biopsy in ovarian cancer, focusing on two key blood-borne DNA sources: cell-free DNA (cfDNA) and DNA carried within extracellular vesicles (EV-DNA). Ovarian cancer causes disproportionately high mortality among gynecologic cancers because the vast majority of cases are not detected until they have reached advanced stages (III or IV), when the disease has spread to the peritoneum and treatment options are severely limited. Liquid biopsy offers a non-invasive alternative to traditional tumor biopsy by detecting cancer-associated biomarkers — including nucleic acids, proteins, and lipids — directly from a blood sample, enabling real-time monitoring of disease progression and treatment response while overcoming the spatial and temporal limitations of tissue sampling. The authors detail how cfDNA and EV-DNA each capture distinct aspects of tumor biology and together provide complementary windows into tumor heterogeneity and mechanisms of treatment resistance. The review concludes with a strong call for further research into the functional roles of these DNA fragments and their integration into multi-omics frameworks to ultimately improve early detection rates and guide personalized therapeutic strategies for ovarian cancer patients.
Extracellular vesicles and circulating nucleic acids
Source →Metastatic disease in conjunctival malignant melanoma: a retrospective analysis of 167 cases.
Sezgin M, et al
Researchers from University Hospital Essen and University Hospital Tübingen in Germany conducted a retrospective study of 167 patients with conjunctival malignant melanoma — a rare but serious cancer affecting the surface of the eye — to identify risk factors for metastatic spread and evaluate the effectiveness of adjuvant therapies following tumor removal. Over an average follow-up period of over six years, nearly half of patients (47.3%) experienced local tumor recurrence, while 24.5% developed metastases, either in regional lymph nodes, distant organs, or both. A key finding was that ruthenium-106 brachytherapy, a form of targeted radiation treatment applied directly to the eye, reduced the risk of blood-borne (haematogenous) distant metastasis by 74% in patients with localized tumors on the bulbar conjunctiva, with only 4 out of 54 treated patients developing distant metastases. The study also found that patients who did not receive any adjuvant therapy after tumor excision had notably higher recurrence rates, underscoring the importance of post-surgical treatment. These findings provide clinicians with strong evidence to support the routine use of adjuvant therapies, particularly brachytherapy, in the management of conjunctival melanoma to improve long-term patient outcomes.
BMJ open ophthalmology
Source →The Gastro-Intestinal Microbiota in Haematology.
Moreno-Mirón JM, et al
A new review published in Acta Haematologica examines the growing body of evidence linking the gut microbiota to blood cell formation and haematological diseases, including clonal haematopoiesis of indeterminate potential (CHIP), leukaemias, and plasma cell neoplasms such as multiple myeloma. The authors describe several biological mechanisms through which gut bacteria may influence blood and immune cell development, including the production of short-chain fatty acids, activation of pattern-recognition receptors, regulation of intestinal barrier integrity, and cytokine-driven inflammation. The review also highlights that common haematological treatments — such as antibiotics, chemotherapy, immunomodulatory drugs, and stem cell transplantation — can significantly alter the composition of the gut microbiome, with downstream effects on infection risk, treatment tolerance, and patient outcomes. While most existing evidence is associative rather than causal, the authors argue that the gut microbiota represents a potentially modifiable factor in haematological disease risk and treatment response. They call for rigorous longitudinal and interventional studies, as well as multi-omics approaches, to establish causality and ultimately enable microbiota-informed strategies in clinical haematology.
Acta haematologica
Source →