There's A New Tool Used To Predict Patient's Survival Time, Learn About It Here

Doctors often tell their patients how long they are likely to live and how well their body is responding to treatments. However, this can be one of the trickiest parts of their job. Now, a new statistical method promises to give accurate information on how long a cancer patient may survive.

According to a report by Tech Times, researchers from UCLA have developed a new tool called Survival Analysis of mRNA Isoform Variation (SURVIV), which shows the measurement uncertainty of mRNA isoform ratio in RNA-sequencing data in order to predict patients' survival time. Researchers also said that this tool should help them evaluate the success of treatments.

The simulation tests were done for six different types of cancer involving different type of malignancies that affect the breast, brain, kidneys, lungs and the ovaries. Researchers spent more than 2 years to develop the SURVIV algorithm by using tissue samples from 2,684 patients. After collecting the said tissues, they compared the survival time approximated by the algorithm with the amount of time the patients had actually survived. The group also included some patients who were still alive.

"[W]e found that isoform-based predictions work consistently better than the conventional gene-based predictions in predicting survival time," lead study author Dr. Yi Xing told Medscape, adding that it could take one to three years for their innovation to be used in clinical settings.

Dr. Yi Xing also said that by using biomedical big data that surrounds the molecular and clinical profiles of cancer, they will be able to identify unique biomarkers that lead cancer prognosis and treatment.

Newsmax reported that by using a metric known as the C-index, the scientists evaluated the performance of the survival predictors, and found out that their new tool performed better than the conventional gene-based predictions across simulation studies for the six cancer types.

Now the team is using SURVIV to a much larger datasets across many more types of cancers in order to develop more reliable isoform-based predictors of patient survival. Dr. Xing explained that they are hoping to discover more isoforms that are consistently being associated with survival in a "pan-cancer" analysis across multiple cancer types.  

This method could be extended to predict other types of patient outcomes, such as response to specific therapies.