The healthcare industry has seen numerous use cases of computers and trained algorithms in the recent past. They have been used in the detection of brain hemorrhages and breast cancer using neural network systems and machine learning models.
Now, scientists at Uppsala University in collaboration with researchers at Lund University, Karolinska Institute, and the Chalmers University of Technology, have employed a computer algorithm in devising a novel treatment for neuroblastoma.
To be specific, neuroblastoma is a potentially life-threatening form of cancer in children, which occurs in specialized nerve cells in the sympathetic nervous system. The paper that has been published in Nature Communications, titled "Integrative discovery of treatments for high-risk neuroblastoma", states that "approximately 50% of children with high-risk neuroblastoma lack effective treatment."
The modus operandi of the team of researchers was as follows. Unlike traditional drug development procedures, the team analyzed genetic and pharmacological data from various European and American hospitals and universities using a smart algorithm. The algorithm then suggested new treatments that could influence the basic mechanisms of neuroblastoma.
Subsequently, the researchers "establish[ed] CNR2 and MAPK8 as promising candidates for the treatment of high-risk neuroblastoma." One of the proposed treatments is based on activating a receptor protein, CNR2 (cannabinoid receptor 2), in the nervous system, to suppress the growth of tumors. Likewise, the mitogen-activated protein kinase 8 (MAPK8) also served the same utility.
We then characterise selected targets by more than 700 RNA profiling experiments in drug-treated neuroblastoma cells and show that interfering with two drug targets, the mitogen-activated protein kinase 8 (MAPK8) and the cannabinoid receptor 2 (CNR2) suppress tumour growth in both zebrafish and mouse xenograft models. Together, these results deepen our understanding of neuroblastoma vulnerabilities and provide a tool for data-guided cancer target discovery.
To test the efficacy of the solution proposed by the algorithm, the new treatments were investigated using cell samples from patients and in animal models. Resultantly, the tests carried out on them showed that "the cancer cells’ survival rate declined, for example, and tumor growth in Zebrafish (Danio rerio) decreased, following treatment with a substance that stimulates CNR2," showing that the proposed solutions were effective.
The team, led by Sven Nelander, who is a senior lecturer at Uppsala University commented on the potential advantages proposed by smart algorithms in the field of cancer research:
"Smart algorithms will be increasingly important in cancer research in the years ahead, since they can help us scientists to find unexpected angles. We’ve already started a major project here in Uppsala, in which several types of cancer in children and adults will be investigated this way. Our hope is that this can result in more unexpected treatment options."
Interestingly, Nelander's team also claims to have generalized the computer algorithm to be applied to other forms of cancer. If you are interested in finding out more, you may study the paper here.