Artificial intelligence has shown considerable superiority over biopsy in grading histological type and grade of some sarcomas, UK researchers have reported.

Data suggests that a newly-developed AI algorithm looks to be at least twice as accurate and could therefore help to better tailor treatment for patients than current methods, according to a study published in The Lancet Oncology,

The new algorithm could assist clinicians with faster diagnosis of disease subtypes, thus facilitating faster access to treatment, the researchers say.

The team at The Royal Marsden NHS Foundation Trust, London, analysed CT scans from 170 patients (median age 63 years; 49% female) treated for leiomyosarcoma and liposarcoma to develop an AI algorithm, which was then validated in 89 patients from centres across Europe and the US (median age 59 years; 52% female).

The model was able to accurately predict histological grade in 82% of those analysed, compared to 44% of those graded via biopsy.

Disease type was accurately identified by AI in 84% of the sarcomas tested, while “the reporting radiologist was not able to offer a diagnosis in 35% of patients and was only able to correctly diagnose 73% of liposarcoma and 43% of leiomyosarcoma,” according to the paper.

The researchers said the findings “provide a foundation for the further development and prospective validation of these models”.

“We’re incredibly excited by the potential of this state-of-the-art technology, which could lead to patients having better outcomes through faster diagnosis and more effectively personalised treatment,” said lead investigator Professor Christina Messiou, Consultant Radiologist at The Royal Marsden.

“As patients with retroperitoneal sarcoma are routinely scanned with CT, we hope this tool will eventually be used globally, ensuring that not just specialist centres – who see sarcoma patients every day – can reliably identify and grade the disease.

“In the future, this approach may help characterise other types of cancer, not just retroperitoneal sarcoma. Our novel approach used features specific to this disease, but by refining the algorithm, this technology could one day improve the outcomes of thousands of patients each year.”

“In the next phase of the study, we will test this model in clinic on patients with potential retroperitoneal sarcomas to see if it can accurately characterise their disease and measure the performance of the technology over time,” added first author Dr Amani Arthur, Clinical Research Fellow at The ICR and Registrar at The Royal Marsden.