Algorithm for bowel-cancer patient care

An algorithm that can predict how long a patient might spend in hospital if they’re diagnosed with bowel cancer will allow healthcare providers to design the best patient care and prioritise resources. That’s according to experts from the University of Portsmouth and the Portsmouth Hospitals University NHS Trust, who have used artificial intelligence and data analytics to predict the length of hospital stay for bowel cancer patients, whether they will be readmitted after surgery, and their likelihood of death over a one or three-month period. Their research used data taken from a database of over 4,000 bowel cancer patients who underwent surgery between 2003 and 2019. It looked at 47 different variables including age, weight, fitness, surgical approaches, and mortality. The insights of consultant surgeon Jim Khan and his colleagues Samuel Stefan and Karen Flashman were complemented by the analytical expertise of Dr Shamsul Masum, under Professor of Intelligent Systems, Prof Adrian Hopgood’s direction. Prof Hopgood plans to work with an even bigger dataset to improve the accuracy of predictions, which is already above 80%.

New PCR test for oral cancer 

Researchers from Queen Mary University of London have developed the first polymerase chain reaction (PCR) test for oral cancer detection. Besides yielding rapid results, the test, called the quantitative Malignant Index Diagnosis System (qMIDS), is cost-efficient and easy to use and could relieve pressure on healthcare services. Similar to the Covid-19 PCR tests, the qMIDS test detects RNA. It requires a tiny 1mm3 tissue biopsy, which can be taken by a general dentist or an oral surgeon. An algorithm calculates and converts a panel of 16 gene mRNA expression levels into a qMIDS index to quantify risk of malignancy for each sample. The new qMIDS assay was recently validated in a UK oral squamous cell carcinoma (OSCC) cohort (n = 282). It was also independently validated in Chinese and Indian OSCC patients. Senior researcher Dr Iain Hutchison, professor of oral and maxillofacial surgery at Queen Mary, said: “qMIDS will help us identify patients with pre-malignancies that will never transform into cancer, so they can be reassured and discharged from hospital review. Patients with high-risk pre-malignancy can have minor surgery to remove the lesion before it has transformed into cancer, thereby curing the patient and saving them major surgery, which in turn reduces health service costs.”

AI system cuts lung cancer diagnosis time

An artificial intelligence system about to be trialled in NHS hospitals in ther UK could halve the time it takes to get a lung cancer diagnosis. The tool harnesses computer power to examine x-rays. An early study found that immediate reading cuts the time taken from the first chest x-ray to CT scan and diagnosis from 60 to 30 days, and AI could aid with this. The AI will only prioritise abnormal chest X-rays, the normal cases follow the usual pathway. The new system spots x-rays with suspicious lung nodules within minutes so that patients can have a more detailed CT scan for a more accurate diagnosis before they go home. The NHS pilot will run at seven hospitals including Nottingham University Hospital Trust, The Royal Marsden NHS Trust in south London, and University College London Hospital.

‘Drug factory’ implants for cancer

Rice University bioengineers  in Texas have shown they can eradicate advanced-stage ovarian and colorectal cancer in mice in as little as six days with a treatment that could be ready for human clinical trials in the US later this year. The researchers used implantable “drug factories” the size of a pinhead to deliver continuous, high doses of interleukin-2, a natural compound that activates white blood cells to fight cancer. 

The drug-producing beads can be implanted with minimally invasive surgery. Each contains cells engineered to produce interleukin-2 that are encased in a protective shell. “We just administer once, but the drug factories keep making the dose every day, where it’s needed until the cancer is eliminated,” said Omid Veiseh, an assistant professor of bioengineering whose lab produced the treatment. “Once we determined the correct dose – how many factories we needed – we were able to eradicate tumours in 100% of animals with ovarian cancer and in seven of eight animals with colorectal cancer.”

Smartphone game boosts cancer research

Cancer Research UK has unveiled Play to Cure: Genes in Space – a world-first mobile phone game in which people across the globe will be able to help scientists unravel gene data to find the answers to some of cancer’s toughest questions. It is available to download for free for anyone with an Android or Apple Smartphone. When playing this fun and interactive spaceship game, people will simultaneously analyse Cancer Research UK’s gene data, highlighting genetic faults which can cause cancer – and ultimately help scientists develop new treatments. Players must guide a fast-paced spaceship safely along a hazard-strewn intergalactic assault course to collect precious material called ‘Element Alpha’. Each time the player steers the spaceship to follow the Element Alpha path, this information is fed back to Cancer Research UK scientists – cleverly providing analysis of variations in gene data. Scientists need this information to work out which genes are faulty in cancer patients – so they can develop new drugs that target them, speeding their progress towards personalised medicine. Each section of gene data will be tracked by several different players to ensure accuracy.