Cancer affects millions of Americans every year and despite progress in detection, surveillance and treatment, it is still the second most common cause of death in the US. Some cancers such as melanoma or brain cancer affect people of every age, ethnicity, and gender. But of the gender-specific cancers, one of the most common and deadly affecting men is prostate cancer which is responsible for 30,000 death in the US each year. The emergence of new more effective therapies has brought a renewed focus on the impact of such therapies on patients’ lives before and after treatment. Here, a new web-based tool using artificial intelligence (nomogram) may help shed some light on the quality of life patients experience after treatment.
This new web-based tool was able to assist in predicting likely outcomes for five years after localized prostate cancer treatment, which consisted of intensity-modulated radiation therapy (IMRT) or radical prostatectomy (RP), and active surveillance. IMRT uses linear accelerators to deliver radiation in the most precise fashion where it simultaneously provides a specific treatment while minimizing the effect on surrounding tissue and cells making this form of therapy especially safe. Radical prostatectomy, which consists of complete excision of the prostate gland and the tissue surrounding it can be performed “openly” through an incision or laparoscopically. Finally, active surveillance is based on the concept that minimally aggressive prostate cancer is unlikely to have an effect on life expectancy. Active surveillance is usually seen as a better option than radiation or surgery when it comes to a low-risk prostate cancer diagnosis.
Taking into account the different treatment options as well as the relative aggressiveness of the tumor, the web-based tool would consider a number of variables including prostate specific-antigen, pre-treatment baseline function, race, and age to produce an expected outcome. As a result, this new tool was able to provide a “personalized” approach to prostate cancer in order to better predict outcomes after treatment, especially measures of quality of life such as urinary incontinence, bowel function, hormonal function, and erectile dysfunction.
Understanding outcomes after cancer therapy is critically important. Using prediction models anchored in artificial intelligence allows to more accurately predict outcomes after therapy and more importantly to personalize such predictions.